WO2018121375A1

WO2018121375A1 - Image matching method and apparatus

Info

Publication number: WO2018121375A1
Application number: PCT/CN2017/117433
Authority: WO
Inventors: 孙鹏; 李建方; 刘琦玉; 兰天; 冯姗姗; 王佳裕
Original assignee: 北京奇虎科技有限公司
Priority date: 2016-12-29
Filing date: 2017-12-20
Publication date: 2018-07-05

Abstract

Provided are an image matching method and apparatus. The method comprises: receiving an image matching request from a client, wherein the image matching request carries key information about the client; according to a pre-set rule, calculating the key information to obtain a first keyword key; matching a queue corresponding to a first key from queues with different keys as indices stored in a database, wherein each queue stores an image recognition document of each image from different clients but with the same key; and reading a matched image recognition document of a queue, and matching each image recognition document according to a pre-set matching logic to obtain an image matching result, and based on the image matching result, correspondingly obtaining a matching result between the present client and other clients. A relationship existing between each image respectively coming from different clients can be embodied by means of the present invention, not only increasing interaction approaches between different clients, but also improving the sociability and interestingness of a client.

Description

Image matching method and device

Technical field

The present invention relates to the field of computer science and technology, and in particular, to an image matching method and apparatus.

Background technique

With the development of Internet technology, many other fields of technology have also been driven to develop. As a result, people's demands for quality of life have become higher and higher. Images are the main source of human access and exchange of information. Therefore, the application of image processing must also involve all aspects of human life. Although people face different problems in their daily lives, there are more and more ways to solve them. However, at this stage, there are still many problems when people process images.

With the continuous development of image processing technology, the field of image processing involves various aspects such as human life and work. The image processing technology in the prior art is more and more mature. By analyzing the feature information of different images, different images can be identified, so that different images can be effectively distinguished.

However, most of the current techniques for image processing in the client are to analyze a single image to obtain some important information in a single image. The processing technique of this image is in the process of processing different images. Independent of each other, and the resulting processing results of the individual images are not necessarily related. Therefore, the existing image processing method is relatively single, which is not conducive to analyzing and acquiring the relationship between multiple images. At the same time, processing a single image is not conducive to interaction between multiple clients.

Therefore, there is a need for a processing method that can conveniently provide users with images that meet their needs.

Summary of the invention

In view of the above problems, the present invention has been made in order to provide an image matching method and apparatus that overcomes the above problems or at least partially solves the above problems.

According to an aspect of the present invention, an image matching method is provided, including:

Receiving an image matching request from a client, where the image matching request carries key information of the client; calculating the key information according to a preset rule to obtain a first keyword key; storing in a database The queues indexed by different keys are matched with the queue corresponding to the first key, wherein each queue stores image recognition files of images from different clients but having the same key; the read matches The images in the queue identify the copy, and match each image recognition copy according to the preset matching logic to obtain the image matching result, and obtain the matching result between the client and other clients according to the image matching result.

According to another aspect of the present invention, an image matching method is further provided, including:

Receiving, by the first client, a first image recognition request carrying the first image and the key information, identifying the first image to obtain a first image recognition copy; receiving the second image from the second client a second image recognition request of the key information, identifying the second image to obtain a second image recognition copy; calculating the key information according to a preset rule, obtaining a keyword key, and establishing a key For the indexed queue, the first image recognition copy and the second image recognition copy are placed in the queue for storage; when the matching request of the first client and the second client is received, according to Key to the queue to read the first image recognition copy and the second image recognition copy; matching the first image recognition copy and the second image recognition copy to obtain the first customer The result of the matching between the terminal and the second client.

According to another aspect of the present invention, there is also provided an image matching apparatus comprising:

The receiving module is adapted to receive an image matching request from the client, where the image matching request carries the key information of the client, and the calculating module is configured to calculate the key information according to a preset rule to obtain a first keyword key; a queue matching module, configured to match a queue corresponding to the first key in a queue indexed by a different key stored in a database, wherein each queue stores a code from a different client, but An image recognition copy of each image having the same key; the image matching module is adapted to read the image recognition copy in the matched queue, and match each image recognition copy according to a preset matching logic to obtain an image matching result, and based on The image matching result is matched by the client and other clients.

The first receiving module is adapted to receive a first image recognition request from the first client that carries the first image and the key information, and identify the first image to obtain a first image recognition copy; the second receiving module, Suitable for receiving a second image recognition request from the second client carrying the second image and the key information, identifying the second image to obtain a second image recognition copy; and a storage module adapted to be preset according to the preset The rule calculates the key information, obtains a keyword key, and establishes a queue indexed by the key, and places the first image recognition copy and the second image recognition copy into the queue for storage; a module, configured to read the first image recognition copy and the second image recognition copy into the queue when receiving a matching request between the first client and the second client; a matching module, And matching the first image recognition copy and the second image recognition copy to obtain a matching result between the first client and the second client.

According to another aspect of the present invention, there is also provided a computer program comprising computer readable code, when said computer readable code is run on a computing device, causing said computing device to perform according to any of the above Image matching method.

According to another aspect of the present invention, there is also provided a computer readable medium storing a computer program as described above.

In the present invention, a plurality of queues are pre-stored in the database, and each queue is indexed by a key. Each queue stores image recognition files of images from different clients but having the same key, that is, a database to the client. The image requested for identification is stored in groups, and the same key is stored in the same queue, which provides a data basis for subsequent matching operations. Further, if the image matching request from the client is received, the key information carried in the matching request is calculated, and the first key is obtained, and then the first key is matched with each queue in the database, and the first key is matched. The queue, that is, the queue indexed by the first key, can obtain each image recognition copy in the queue indexed by the first key. Then, each image recognition copy in the queue indexed by the first key is read, and each image recognition copy is matched according to a preset matching logic to obtain an image matching result. Because, and the image recognition copy is from different clients, the matching result between the client and other clients is obtained based on the image matching result. The matching result can reflect the relationship between the images from different clients, which not only increases the interaction between different clients, but also improves the sociality and fun of the client. In addition, for users who use the client, not only can the interaction between images owned by different users be increased, but also communication and communication between different users can be enhanced. In addition, the matching request carries the key information, and the key information is used as the key information of the matching queue, so that the matching operation can be limited to the image recognition copy of each image from different clients but having the same key, and the other clients are guaranteed to send to The security of the image of the server makes the user's private information not easily leaked.

The above description is only an overview of the technical solutions of the present invention, and the above-described and other objects, features and advantages of the present invention can be more clearly understood. Specific embodiments of the invention are set forth below.

DRAWINGS

Various other advantages and benefits will become apparent to those skilled in the art from a The drawings are only for the purpose of illustrating the preferred embodiments and are not to be construed as limiting. Throughout the drawings, the same reference numerals are used to refer to the same parts. In the drawing:

1 is a flow chart showing an image matching method according to an embodiment of the present invention;

2 is a process flow diagram of a method for generating an image recognition copy in a queue according to an embodiment of the present invention;

FIG. 3 is a flow chart showing an image matching method according to another embodiment of the present invention; FIG.

4 is a flow chart showing an image matching method according to another embodiment of the present invention;

FIG. 5 is a flow chart showing an image matching method according to another embodiment of the present invention; FIG.

FIG. 6 is a block diagram showing the structure of an image matching apparatus according to an embodiment of the present invention; FIG.

FIG. 7 is a block diagram showing another structure of an image matching apparatus according to an embodiment of the present invention; FIG.

FIG. 8 is a block diagram showing the structure of an image matching apparatus according to another embodiment of the present invention; FIG.

FIG. 9 is a block diagram showing the structure of an image matching apparatus according to another embodiment of the present invention; FIG.

Figure 10 is a schematic block diagram showing a computing device for performing an image matching method in accordance with the present invention;

Fig. 11 schematically shows a storage unit for holding or carrying program code implementing the image matching method according to the present invention.

detailed description

The invention is further described below in conjunction with the drawings and specific embodiments.

Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While the embodiments of the present invention have been shown in the drawings, the embodiments Rather, these embodiments are provided so that this disclosure will be more fully understood and the scope of the disclosure will be fully disclosed.

To solve the above technical problem, an embodiment of the present invention provides an image matching method, which is applied to a server. FIG. 1 is a schematic flow chart of an image matching method according to an embodiment of the present invention. Referring to FIG. 1, the method includes at least steps S102 to S110.

Step S102: Receive an image matching request from a client, where the image matching request carries key information of the client.

Step S104: Perform calculation on the key information according to a preset rule to obtain a first keyword key;

Step S106: Matching queues corresponding to the first key in the queues indexed by different keys stored in the database, wherein each queue stores image recognition of images from different clients but having the same key Copywriting

Step S108: Read the image recognition copy in the matched queue, and match each image recognition copy according to the preset matching logic to obtain an image matching result, and obtain a matching result between the client and other clients according to the image matching result. .

In the embodiment of the present invention, a plurality of queues are pre-stored in the database, and each queue is indexed by a key. Each queue stores image recognition files of images from different clients but having the same key, that is, database pairs. The images identified by the client request are stored in groups, and the same key is stored in the same queue, which provides a data foundation for subsequent matching operations. Further, if the image matching request from the client is received, the key information carried in the matching request is calculated, and the first key is obtained, and then the first key is matched with each queue in the database, and the first key is matched. The queue, that is, the queue indexed by the first key, can obtain each image recognition copy in the queue indexed by the first key. Then, each image recognition copy in the queue indexed by the first key is read, and each image recognition copy is matched according to a preset matching logic to obtain an image matching result. Because, and the image recognition copy is from different clients, the matching result between the client and other clients is obtained based on the image matching result. The matching result can reflect the relationship between the images from different clients, which not only increases the interaction between different clients, but also improves the sociality and fun of the client. In addition, for users who use the client, not only can the interaction between images owned by different users be increased, but also communication and communication between different users can be enhanced. In addition, the matching request carries the key information, and the key information is used as the key information of the matching queue, so that the matching operation can be limited to the image recognition copy of each image from different clients but having the same key, and the other clients are guaranteed to send to The security of the image of the server makes the user's private information not easily leaked.

It should be noted that the image recognition copy in the queue is that the server recognizes the image uploaded by each client in advance, and obtains a corresponding image recognition request. Specifically, FIG. 2 shows a process flow diagram of a method for generating an image recognition copy in a queue according to an embodiment of the present invention. Referring to FIG. 2, the method includes at least steps S202 to S208:

Step S202: Receive an image recognition request from a client, where the image recognition request carries an image to be identified and key information provided by the client.

Step S204: Perform image recognition on the image carried in the image recognition request to obtain an image recognition result.

Step S206, calculating key information of the current image identification according to the preset rule, to obtain a corresponding key;

Step S208, establishing a queue indexed by the key in the database for the key, and placing the image recognition copy into the queue.

After the client has uploaded the image, the client may initiate the image matching request mentioned in step S102 above to find other clients that match the client. In an embodiment of the present invention, after receiving the image matching request from the client, the timing may be started from the time when the matching request is received, thereby obtaining the timing of the matching process. Due to the complexity of different image content, if the image content is more complicated, it takes more image recognition and image matching time. In addition, in the process of image transmission, there may be unexpected situations such as file loss or network interruption. At this time, the time of the image matching process can be effectively controlled by timing the matching process, and differently The timing of the time takes a corresponding solution to improve the matching efficiency of image matching.

Specifically, before reading each image recognition copy in the queue according to the first key to the first key index, it may be determined whether the time duration from the time of the matching request exceeds the rated duration of the current match, and if so, The matching operation is suspended, and if not, the reading operation of each image recognition copy is performed. In this embodiment, the rated duration of the current matching may be preset, for example, the rated duration is 2 seconds, and if the matching request request is received for more than 2 seconds, the matching operation is suspended, if received from the receiving When the matching request time starts no more than 2 seconds, each image recognition copy is read according to the first key to the queue indexed by the first key. If the timeout expires, the matching operation is directly suspended, and an unsuccessful error code is sent to the client to inform the client that the matching operation has ended.

Continuing to refer to step S108 above, in an embodiment of the present invention, in order to ensure that the content read from the queue is valid and complete, and the subsequent matching operation can be completed, when the image recognition copy is read from the queue, The image recognition copy in the queue is counted, the counting result is obtained, and the counting result is judged correspondingly. If the counting result is greater than or equal to 2, at least 2 image recognition files have been stored in the verification queue, and the image recognition copy matching operation can be performed. Then, in this example, the matching step of each image recognition scheme is continued, and if the counting result is less than 2, It may be found that only one image recognition copy may be stored in the queue, and even no image recognition copy may be stored in the queue, and the matching operation cannot be performed, but at this time, the rated duration of the match is not exceeded yet, and it is still valid. During the matching time, then in this case, the image recognition copy in the queue can be read again at a specified time interval. An image recognition file of less than 2 appears in the queue, which may be because the server has not completed the recognition operation of other images, or the image recognition copy may be deleted during the storage process due to exceeding the queue storage time, or may be data loss.

In this embodiment, the interval time may be set to an interval of 1 ms. Of course, other interval times may be set, such as 0.9 ms, 1.5 ms, etc., and the value of the specific interval time may be set according to actual conditions, in the embodiment of the present invention. There is no limit to this.

Referring to step S108 above, in an embodiment of the present invention, each image recognition copy is matched, and the matching process may be: first, performing weighting calculation on each image recognition copy to obtain a weight value of each image recognition copy. . Then, the weighted values of each image recognition copy are compared to obtain an image matching result, and the image matching result is used as a matching result between the client and other clients.

In this embodiment, if each image includes a plurality of main lines, the server needs to first draw the main lines of each image, and the following describes the drawing process of the main lines of the image.

First, the edge detection algorithm can be used to detect the edge of each line in the image, and the detection result of the edge detection algorithm is obtained. For example, the Canny operator can be used to detect the edge of the image. The essence of the Canny operator to detect the edge is to determine the maximum value of the signal function to determine the pixel edge of the image.

Further, an adaptive contour extraction operation of the line is performed on the detection result map obtained according to the edge detection algorithm, and the line outline of each line is obtained. In the adaptive contour extraction operation, it is required to output a binarized image, that is, only two images of black and white, one gray represents the edge, and the other gray represents the background, and the deeper processing of the edge information can be performed. Make the outline of the line clearer.

After the initial outline of the line, there may be some contours with stray interference. In order to further obtain a complete and clear outline of the main line, it is also possible to filter the line outlines of the acquired lines to obtain a line outline that satisfies the main line standard.

Wherein, when the line contours of the acquired lines are filtered, the values of the pixel points of each line contour can be calculated. After determining the value of the pixel point of each line outline, the line outline in which the number of pixel points is less than N is deleted. In the embodiment of the present invention, N takes a value of 20. Therefore, the value of the pixel point of each line outline in the image obtained at this time is not less than 20. After the above steps, if the ratio of the pixel points of each line contour to the total pixel point of the image is higher than the set ratio, the edge detection algorithm is adjusted, and the edge of each line in the image is re-detected by the new edge detection algorithm. Until you get a line outline that meets the main line criteria. In the embodiment of the present invention, the set ratio is 3%. That is to say, when the value of the pixel point of each line outline is higher than 3% of the total pixel point of the image, the edge detection algorithm is adjusted, and the edge of each line in the image is re-detected by using a new edge detection algorithm until Get the line outline that meets the main line standard. It should be noted that the value of N is 20, and the value of the set ratio is 3%, which is only an example of the embodiment of the present invention. The embodiment does not specifically limit the specific value of N and the set ratio by specific circumstances. For example, N may be 30 or 40, and the set ratio may be 4% or 5%.

Finally, the contour of the line is determined according to the selected contour of the line, and then the contour of the line in the area is merged to draw the corresponding main line.

The drawing method of the above main line is mainly based on the principle that the image is usually provided with sufficient information by the main line, and the more complicated various images are analyzed and processed, and the main lines are extracted to make the content of the image simple. It is clear, clear, and obvious, so that the amount of image information can be better extracted, and the important information in the image caused by the blurring of the image itself is avoided.

After drawing the main lines of each image, you can separately set the weights and weight ratios for the copy information of each main line, and then weight the image recognition files according to the weights and weight ratios of each main line. And further, the total weighting value of each image recognition copy is obtained.

In the following, a weighting calculation process of an image recognition copy is introduced in a specific embodiment. In this embodiment, the image recognition copy is an image recognition copy related to a flower. In this example, the client provides a rose image, and other clients. Lily images and rose flower images are provided, and the same key information 1234 is used by all three clients.

For example, since different flowers have different overall contours and stamens and can clearly express the characteristics of the flowers, the outlines of the flowers and the length of the stamen can be used as a main line.

After the server respectively recognizes the image recognition copy of the rose, the image recognition copy of the lily, and the image recognition copy of the rose, the three image recognition files are placed in the same queue, and the index of the queue is the key information 1234. The calculated key = 01.

Then, upon receiving the matching request initiated by the client, the server reads the image recognition copy of the rose, the image recognition copy of the lily, and the image recognition copy of the rose from the queue of key=01, and sets the outline of each flower. The weight ratios of the length of the flower and the length of the stamen are 60% and 40%, respectively.

First, the image recognition copy of the rose and the image recognition copy of the lily are matched. In the copy information of the lily image, the weights of the outline of the flower and the length of the stamen are respectively set to 70 and 30, and the total weighted value of the lily image recognition copy is calculated as S1=70*60%+30*40%=54. In the copy information of the rose image, the weights of the outline of the flower and the length of the stamen are respectively set to 40 and 10, and the total weighted value of the rose image recognition copy is calculated as S2=40*60%+10*40%=28. Comparing the total weighting value S1 and the total weighting value S2 can be obtained, and the difference between the total weighting values 54 and 28 between the two is large, that is, the difference between the lily image and the rose image is large.

In addition, combined with real life, there is indeed a big difference between lily and rose. The petals of the lily are long and the petals have a layer of petals. The petals of the rose are similar to a circle, and the flower has multiple layers of petals. The lily of the open lily is more pronounced, and the length of the lily is longer, while the flower of the open rose is surrounded by the petals of the layer, which is not obvious, and the length of the rose is shorter.

Secondly, the image recognition copy of the rose and the image recognition copy of the rose flower are matched. In the copy information of the rose flower image, the weights of the outline of the flower and the length of the stamen are respectively set to 30 and 20, and the total weighted value of the rosette image recognition copy is calculated as S2=30*60%+20*40%=26. In the copy information of the rose image, the weights of the outline of the flower and the length of the stamen are respectively set to 40 and 10, and the total weighted value of the rose image recognition copy is calculated as S2=40*60%+10*40%=28. Comparing the total weighting value S2 with the total weighting value S3 can be obtained, and the difference between the total weighting values 28 and 26 between the two is small. Therefore, there is little difference between the rose flower image and the rose image.

After obtaining the image matching result, step S108 also mentions that the matching result of the client and other clients can be obtained correspondingly based on the image matching result. In order to ensure that the client and the image can be correspondingly identified, in particular, the queue provided by the embodiment of the present invention further stores an uploaded image to obtain a client ID of the image recognition copy, and correspondingly matches the client ID with the image recognition copy of the uploaded image. storage. When the image matching result is obtained, the corresponding client ID is reversely searched according to each image recognition copy participating in the matching operation, and then the client corresponding to the client ID is confirmed. Therefore, the image recognition copy and the client can be correspondingly formed, so that the image matching result can reflect the matching result between the clients. Therefore, the image matching result can be displayed as a match between the client and other confirmed clients. result.

As mentioned above, there may be more than 2 image recognition files in the queue. Therefore, there must be more than two matching results between the client and other clients. In this case, to improve the statistics of the data, increase the sociality of the client. According to the matching result of the client and other clients, another client matching the client with the highest degree can be selected, and the selected other client and the corresponding matching result are returned to the client.

Embodiments of the present invention provide another image matching method, which is applied to a server. FIG. 3 is a schematic flowchart diagram of an image matching method according to an embodiment of the present invention. Referring to FIG. 3, the method includes at least steps S302 to S310.

Step S302: Receive a first image recognition request from the first client that carries the first image and the key information, and identify the first image to obtain a first image recognition copy.

In this step, the key information may be a password input by the user, and the composition form of the password, the length of the password, and the like may be set by the user. For example, the key information may be a digital password such as "1234", or an alphanuclear password such as "abcd", or a combination of any characters. The specific form of the key information is not limited in the embodiment of the present invention.

Step S304, receiving a second image recognition request from the second client that carries the second image and the key information, and identifying the second image to obtain a second image recognition copy.

In this step, only the key information sent by the second client is consistent with the key information sent by the first client in step S302. In the subsequent step, the server can send the received first client. The first image and the second image sent by the second client are placed in the same storage space after the end of the identification for subsequent matching processing.

Step S306: Calculate the key information according to the preset rule, obtain a keyword key, and establish a queue indexed by the key, and store the first image recognition copy and the second image recognition copy in a queue.

In this step, the keyword key is a corresponding calculation result obtained by calculating the key information by a preset rule. For example, the keyword key may be a calculated machine code composed of “0” and “1”, and also It can be a string of random numbers, and can also be any other form of string. The embodiment of the present invention does not specifically limit the form of the calculation result. The function of the keyword key is mainly to perform related indexing on the queue established by the server. In the queue, the keyword key can correspond to two image recognition files related to one key information.

In addition, the queue mentioned in the step may be in the form of a table, and the keyword key may correspond to a line of content in the table, a column of content, or content in a certain cell, and according to the keyword key, the specified in the table may be found. The content, that is, the index relationship is established between the keyword key and the queue. The storage content of the queue can be set to be time-sensitive. If it is longer than one time, it can be deleted from the queue to ensure the timeliness and validity of the information. In this embodiment, the queue can be kept in the specified database, and multiple queues can be saved in the database at the same time.

Step S308, when receiving the matching request of the first client and the second client, reading the first image recognition copy and the second image recognition copy according to the key to the queue.

In this step, when the server receives the matching request sent by any one of the first client and the second client, the server may obtain the keyword key according to the matching request, and then use the keyword key to index from the queue. The first image recognition copy and the second image recognition copy corresponding to the keyword key are saved in the queue.

Step S310, matching the first image recognition copy and the second image recognition copy to obtain a matching result between the first client and the second client.

In the embodiment of the present invention, the server respectively receives image related information (including image and key information) from the first client and the second client, so that two image recognitions can be separately generated according to image related information from different clients. Copywriting. Moreover, the server may further calculate the received key information according to a preset rule to obtain a keyword key, and at the same time, use the keyword key to establish a key-indexed queue, and put two image recognition files into the queue. In storage. When the server receives the matching request of the first client and the second client, the corresponding two image recognition files are read according to the keyword key to the queue, and then the read two image recognition files are matched to obtain Corresponding matching results. The matching result can reflect the relationship between the images from the first client and the second client respectively, which not only increases the interaction between different clients, but also improves the sociality and interest of the client. In addition, for users who use the client, not only can the interaction between images owned by different users be increased, but also communication and communication between different users can be enhanced.

Further, the server receives the key information input by the user of the first client and the user of the second client, respectively, so that the first image and the second image received by the server are not maliciously stolen, thereby ensuring the client. The security of the image sent to the server, so that the user's private information will not be easily leaked.

The embodiment of the invention further provides a method for reading an image recognition copy, which is used for reading a corresponding image recognition copy from a queue in which an image recognition copy is pre-stored. Referring to FIG. 4, the method may at least include steps S402 to S410.

Step S402, receiving a matching request of the first client and/or the second client.

In this step, the matching request carries the key information. For the specific form of the key information, refer to the foregoing embodiment. The key information can be entered by a user of the first client and/or the second client.

Step S404, parsing the matching request to obtain the key information carried therein.

Step S406: Calculate the key information according to the preset rule, and obtain a key corresponding to the key information.

In this step, by calculating the key information according to the specified preset rule, a keyword key corresponding to the key information can be obtained, and the keyword key is used to index the corresponding image recognition copy from the queue.

In step S408, the key corresponding to the key information is matched with each queue in the database, and a matching queue indexed by the key is found.

Since the index relationship between the keyword key and the specified content in the queue has been established in the queue, the matching is performed by using the keyword key calculated in step S406 and the keyword key in each queue in the database. When the keyword key is matched, the matching queue with the key index can be found, and the image recognition copy in the corresponding queue is found from the queue according to the saved index relationship.

Step S410, reading the first image recognition copy and the second image recognition copy into the queue.

By utilizing the key information input by the user, it can be ensured that the first image and the second image received by the server are not maliciously stolen, thereby ensuring the security of the image sent by the client to the server, so that the private information of the user is not Will be easily leaked.

In order to more clearly illustrate the solution of the embodiment of the present invention, the method for image matching of the present invention is introduced in a specific embodiment, and the method includes at least steps S502 to S522.

First, step S502 is performed to receive a first image recognition request from the first client that carries the first image and the key information, and call the image processing engine to identify the first image to obtain a first image recognition copy.

Then, step S504 is executed to receive a second image recognition request from the second client carrying the second image and the key information, and call the image processing engine to identify the second image to obtain a second image recognition copy. After the first image recognition copy and the second image recognition copy are obtained, step S506 is performed.

Step S506: Calculate the key information according to the preset rule, obtain a keyword key, and establish a queue indexed by the key, and store the first image recognition copy and the second image recognition copy in a queue, and then perform step S508. .

In step S506, the file expiration time of the first image recognition copy and the second image recognition copy may also be set, and the image recognition copy is correspondingly performed by monitoring the time stored in the queue by the image recognition copy and the second image recognition copy. Processing. Specifically, if the first image recognition copy and the second image recognition copy are stored in the queue for more than a set copy time, the first image recognition copy and the second image recognition copy are automatically invalidated. For example, if the file expiration time is set to 2 days, if the image recognition copy is stored in the queue for more than 2 days, the server can automatically delete the expired image recognition copy, or can also identify the expired image recognition copy as unavailable. status. Of course, the image recognition copy may be subjected to other processing, which is not specifically limited in the embodiment of the present invention.

Step S508, when receiving the matching request of the first client and the second client, starting timing with the receiving time of the matching request as a starting point, obtaining the time duration of the matching process, and then performing step S510.

In step S510, it is determined whether the timing duration exceeds the rated duration of the current match. If yes, step S522 is executed to end the current matching. If no, step S512 is executed to count the image recognition copy in the queue to obtain a counting result. After the counting result is obtained, step S514 is performed.

In step S514, it is determined whether the counting result is 2. If yes, go to step S516, if no, go to step S518.

Step S516, reading the first image recognition copy and the second image recognition copy according to the key to the queue, and executing step S520.

Step S518, the image recognition copy in the queue is read again at the specified time interval. Further, the above S510 is executed again.

Step S520, matching the first image recognition copy and the second image recognition copy to obtain a matching result between the first client and the second client. Step S522 is executed to end the current match.

Referring to steps S514 to S518 above, if it is determined that the count result of the image recognition copy in the queue is 2, and the first image recognition copy and the second image recognition copy have been stored in the proof queue, then this example continues to perform each image recognition. In the matching step of the scheme, if the counting result is less than 2, it may be found that only the first image recognition copy or the second image recognition copy may be stored in the queue, and even the image recognition copy may not be stored in the queue, and the matching operation cannot be performed. However, at this time, the rated duration of this match has not been exceeded, and it is still in the valid matching time. At this time, the image recognition copy in the queue can be read again at this time. An image recognition file of less than 2 appears in the queue. It may be because the server has not completed the recognition operation of one or two images. It may also be that the image recognition copy is deleted during the storage process due to exceeding the queue storage time, or it may be data loss. And other reasons.

Based on the same inventive concept, an embodiment of the present invention further provides an image matching apparatus for supporting an image matching method provided by any one of the above preferred embodiments or a combination thereof. FIG. 6 is a block diagram showing the structure of an image matching apparatus according to an embodiment of the present invention. Referring to Figure 6, the device includes at least:

The receiving module 310 is adapted to receive an image matching request from the client, where the image matching request carries the key information of the client;

The calculation module 320 is coupled to the receiving module 310, and is configured to calculate the key information according to a preset rule to obtain a first keyword key;

The queue matching module 330 is coupled to the computing module 320, and is adapted to match a queue corresponding to the first key in a queue indexed by a different key stored in a database, where each queue stores, from different clients, However, the image recognition copy of each image having the same key;

The image matching module 340 is coupled to the queue matching module 330, and is adapted to read the image recognition copy in the matched queue, and match each image recognition copy according to the preset matching logic to obtain an image matching result, and based on the image matching result. Corresponding to the matching results between this client and other clients.

In a preferred embodiment, referring to FIG. 7, the image matching apparatus may further include:

The timing module 350 is coupled to the receiving module 310, and is adapted to receive the image matching request from the client, and start timing according to the receiving time to obtain the time duration.

In a preferred embodiment, the timing module 350 is further adapted to determine whether the timing duration exceeds the rated duration of the current match before reading the image recognition copy in the matched queue; if yes, the termination The sub-matching operation; if not, triggering the image matching module 340 to perform a reading operation of each image-recognition copy.

The sending module 360 is coupled to the timing module 350, and is configured to send an unsuccessful error code to the client after the current matching operation is suspended.

In a preferred embodiment, the image matching module 340 can also be adapted to:

Counting the image recognition copy in the matched queue to obtain a counting result; if the counting result is greater than or equal to 2, performing a matching step of each image recognition scheme; if the counting result is less than 2, reading the interval at a specified time again Take the image recognition copy in the matching queue.

In a preferred embodiment, the image matching module 340 is further adapted to read the image recognition copy in the matched queue again at intervals of 1 ms.

Each image recognition copy is weighted to obtain a weighted value of each image recognition copy; the weighted value of each image recognition copy is compared to obtain an image matching result. In a preferred embodiment, the image matching module 340 is further adapted to: when each image includes a plurality of main lines, separately set a weight and a weight ratio for the copy information of each main line; and weight each image recognition copy Calculate to obtain the total weighted value of each image recognition copy.

The queue also stores an uploaded image to obtain a client ID of the image recognition copy, and stores the client ID corresponding to the image recognition copy of the uploaded image; when the image matching result is obtained, according to each image recognition participating in the matching operation The copy file reversely finds the corresponding client ID, and then confirms the client corresponding to the client ID; the image matching result is displayed as a matching result between the client and other confirmed clients.

The selection module 370 is coupled to the image matching module 340, and is adapted to obtain a matching result between the client and other clients based on the image matching result, and select the highest matching degree with the client according to the matching result between the client and the other client. Another client; return another client selected and the corresponding matching result to the client.

The document identification module 380 is coupled to the queue matching module 330, and is adapted to generate an image recognition copy in each queue by receiving at least two images from the client, and the at least two images carry the same key information; Performing image recognition on the at least two images to obtain an image recognition result; calculating key information of the current image recognition according to the preset rule to obtain a corresponding key; establishing the key in the database for the key The queue is indexed and each image recognition copy is placed in this queue.

The embodiment of the present invention further provides an image matching device. FIG. 8 is a schematic structural diagram of an image matching device according to an embodiment of the present invention. Referring to FIG. 8, the image matching device 400 includes at least a first receiving module 410 and a second. The receiving module 420, the storage module 430, the reading module 440, and the matching module 450.

The functions of the components or devices of the image matching apparatus 400 and the connection relationship between the parts of the image matching apparatus 400 of the embodiment of the present invention will now be described:

The first receiving module 410 is adapted to receive a first image recognition request from the first client that carries the first image and the key information, and identify the first image to obtain a first image recognition copy;

The second receiving module 420 is coupled to the first receiving module 410, and is adapted to receive a second image from the second client that carries the second image and the key information, and identify the second image to obtain a second image recognition copy. ;

The storage module 430 is coupled to the second receiving module 420, and is adapted to calculate key information according to a preset rule, obtain a keyword key, and establish a queue indexed by the key, and identify the first image and the second image. Identify the copy and put it in the queue for storage;

The reading module 440 is coupled to the storage module 430, and is adapted to read the first image recognition copy and the second image recognition copy into the queue when receiving the matching request of the first client and the second client;

The matching module 450 is coupled to the reading module 440 and is adapted to match the first image recognition copy and the second image recognition copy to obtain a matching result between the first client and the second client.

Another embodiment of the present invention further provides an image matching device. FIG. 9 is a schematic structural diagram of an image matching device according to another embodiment of the present invention. Referring to FIG. 9, the image matching device 400 includes the components shown in FIG. In addition to the module, a timing module 510 can also be included.

The timing module 510 is coupled to the reading module 440, and is adapted to start timing when the reading module 440 receives the matching request of the first client and the second client, starting from the receiving time of the matching request, and obtaining the timing of the matching process. duration.

In an embodiment of the invention, the timing module 510 is further adapted to determine whether the chronograph duration exceeds the rated duration of the current match before the reading module 440 reads the first image recognition copy and the second image recognition copy in the queue. If yes, the matching operation is aborted; if not, the reading operation of each image recognition copy is performed.

In an embodiment of the invention, the reading module 440 is further adapted to count the image recognition copy in the queue to obtain a counting result; if the counting result is 2, perform a matching step of each image recognition scheme; if the counting result is less than 2. The image recognition copy in the queue is read again at the specified time interval.

In an embodiment of the invention, the reading module 440 is further adapted to read the image recognition copy in the queue again at intervals of 1 ms.

In an embodiment of the present invention, the reading module 440 is further adapted to first parse the matching request to obtain the key information carried therein, and calculate the key information according to a preset rule to obtain a key corresponding to the key information, and then Matching the key corresponding to the key information with each queue in the database, finding a matching queue indexed by the key, and finally reading the first image recognition copy and the second image recognition copy into the queue.

In an embodiment of the present invention, the matching module 450 is further configured to perform weighting calculation on the first image recognition copy and the second image recognition copy respectively, and obtain weight values of the first image recognition copy and the second image recognition copy, and compare The weighted values of the first image recognition copy and the second image recognition copy obtain an image matching result, and the image matching result is used as a matching result between the first client and the second client.

In an embodiment of the present invention, the matching module 450 is further adapted to separately set a weight and a weight ratio for the copy information of each main line when the first image and/or the second image includes a plurality of main lines, according to each main line The weight and the weight ratio respectively perform weighting calculation on each image recognition copy to obtain a total weight value of each image recognition copy.

The image matching method and apparatus provided by the embodiments of the present invention can achieve the following beneficial effects:

In the description provided herein, numerous specific details are set forth. However, it is understood that the embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques are not shown in detail so as not to obscure the understanding of the description.

Similarly, the various features of the invention are sometimes grouped together into a single embodiment, in the above description of the exemplary embodiments of the invention, Figure, or a description of it. However, the method disclosed is not to be interpreted as reflecting the intention that the claimed invention requires more features than those recited in the claims. Rather, as the following claims reflect, inventive aspects reside in less than all features of the single embodiments disclosed herein. Therefore, the claims following the specific embodiments are hereby explicitly incorporated into the embodiments, and each of the claims as a separate embodiment of the invention.

Those skilled in the art will appreciate that the modules in the devices of the embodiments can be adaptively changed and placed in one or more devices different from the embodiment. The modules or units or components of the embodiments may be combined into one module or unit or component, and further they may be divided into a plurality of sub-modules or sub-units or sub-components. In addition to such features and/or at least some of the processes or units being mutually exclusive, any combination of the features disclosed in the specification, including the accompanying claims, the abstract and the drawings, and any methods so disclosed, or All processes or units of the device are combined. Each feature disclosed in this specification (including the accompanying claims, the abstract and the drawings) may be replaced by alternative features that provide the same, equivalent or similar purpose.

In addition, those skilled in the art will appreciate that, although some embodiments described herein include certain features that are included in other embodiments and not in other features, combinations of features of different embodiments are intended to be within the scope of the present invention. Different embodiments are formed and formed. For example, in the following claims, any one of the claimed embodiments can be used in any combination.

The various component embodiments of the present invention may be implemented in hardware, or in a software module running on one or more processors, or in a combination thereof. Those skilled in the art will appreciate that some or all of the functionality of some or all of the components of the image matching device in accordance with embodiments of the present invention may be implemented in practice using a microprocessor or digital signal processor (DSP). The invention can also be implemented as a device or device program (e.g., a computer program and a computer program product) for performing some or all of the methods described herein. Such a program implementing the invention may be stored on a computer readable medium or may be in the form of one or more signals. Such signals may be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.

For example, Figure 10 illustrates a block diagram of a computing device in which an image matching method in accordance with the present invention can be implemented. The computing device conventionally includes a processor 1010 and a computer program product or computer readable medium in the form of a memory 1020. The memory 1020 may be an electronic memory such as a flash memory, an EEPROM (Electrically Erasable Programmable Read Only Memory), an EPROM, a hard disk, or a ROM. Memory 1020 has a storage space 1030 that stores program code 1031 for performing any of the method steps described above. For example, the storage space 1030 storing program code may store respective program codes 1031 for implementing various steps in the above methods, respectively. The program code can be read from or written to one or more computer program products. These computer program products include program code carriers such as hard disks, compact disks (CDs), memory cards or floppy disks. Such computer program products are typically portable or fixed storage units as shown in FIG. The storage unit may have storage segments, storage spaces, and the like that are similarly arranged to memory 1020 in the computing device of FIG. The program code can be compressed, for example, in an appropriate form. In general, the storage unit comprises computer readable program code 1031' storing program steps for performing the method according to the invention, ie program code readable by a processor such as 1010, when the program code is run by the computing device, The computing device is caused to perform the various steps in the methods described above.

"an embodiment," or "an embodiment," or "an embodiment," In addition, it is noted that the phrase "in one embodiment" is not necessarily referring to the same embodiment.

It is to be noted that the above-described embodiments are illustrative of the invention and are not intended to be limiting, and that the invention may be devised without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as a limitation. The word "comprising" does not exclude the presence of the elements or steps that are not recited in the claims. The word "a" or "an" The invention can be implemented by means of hardware comprising several distinct elements and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means can be embodied by the same hardware item. The use of the words first, second, and third does not indicate any order. These words can be interpreted as names.

In addition, it should be noted that the language used in the specification has been selected for the purpose of readability and teaching, and is not intended to be construed or limited. Therefore, many modifications and changes will be apparent to those skilled in the art without departing from the scope of the invention. The disclosure of the present invention is intended to be illustrative, and not restrictive, and the scope of the invention is defined by the appended claims.

Claims

An image matching method includes:

Receiving an image matching request from a client, where the image matching request carries key information of the client;

Calculating the key information according to a preset rule to obtain a first keyword key;

Queues corresponding to the first key are matched in the queues indexed by different keys stored in the database, wherein each queue stores image recognition files of images from different clients but having the same key;

The image recognition copy in the matched queue is read, and each image recognition copy is matched according to the preset matching logic, and the image matching result is obtained, and the matching result between the client and other clients is obtained according to the image matching result.
The method according to claim 1, wherein after receiving the image matching request from the client, the method further comprises: starting timing with the receiving time as a starting point, and obtaining a timing duration of the matching process.
The method according to claim 1 or 2, wherein before the reading the image recognition copy in the matched queue, the method further comprises:

Determining whether the timing duration exceeds the rated duration of the current match;

If yes, the current matching operation is aborted;

If not, a read operation of each image recognition copy is performed.
The method according to any one of claims 1-3, wherein after the suspending the matching operation, the method further comprises: sending an error code that the matching is unsuccessful to the client.
The method according to any one of claims 1 to 4, wherein the performing a read operation of each image recognition copy and matching each image recognition copy according to a preset matching logic comprises:

Counting the image recognition copy in the matched queue to obtain a counting result;

If the counting result is greater than or equal to 2, performing a matching step of each image recognition scheme;

If the count result is less than 2, the image recognition copy in the matched queue is read again at the specified time interval.
The method according to any one of claims 1 to 5, wherein the interval specifies a time to read the image recognition copy in the matched queue again, including: reading the image recognition copy in the matched queue again at intervals of 1 ms .
The method according to any one of claims 1-6, wherein the matching of each image recognition copy according to a preset matching logic to obtain an image matching result includes:

Performing a weighting calculation on each image recognition copy to obtain a weighted value of each image recognition copy;

The weighted values of each image recognition copy are compared to obtain an image matching result.
The method according to any one of claims 1 to 7, wherein the weighting calculation is performed on each image recognition copy to obtain weighting values of each image recognition copy, including:

When each image includes a plurality of main lines, weights and weight ratios are separately set for the copy information of each main line;

Each image recognition copy is weighted to obtain a total weighted value of each image recognition copy.
The method according to any one of claims 1-8, wherein the matching result between the client and other clients is obtained according to the image matching result, including:

The queue also stores an uploading image to obtain a client ID of the image recognition copy, and stores the client ID corresponding to the image recognition copy of the uploaded image;

When the image matching result is obtained, the corresponding client ID is reversely searched according to each image recognition copy participating in the matching operation, thereby confirming the client corresponding to the client ID;

The image matching result is displayed as a match result between the client and other clients confirmed.
The method according to any one of claims 1 to 9, wherein after the matching result of the client and the other client is obtained correspondingly based on the image matching result, the method includes:

According to the matching result of the client and other clients, select another client that has the highest matching degree with the client;

Return the selected other client and the corresponding matching result to the client.
The method according to any one of claims 1 to 10, wherein the image recognition copy in each queue is generated as follows:

Receiving an image recognition request from a client, wherein the image recognition request carries an image to be identified and key information provided by the client;

Performing image recognition on the image carried in the image recognition request to obtain an image recognition result;

Performing calculation on the key information of the current image according to the preset rule to obtain a corresponding key;

A queue indexed by the key is created in the database for the key, and the image recognition copy is placed in the queue.
An image matching method includes:

Receiving, by the first client, a first image recognition request carrying the first image and the key information, and identifying the first image to obtain a first image recognition copy;

Receiving, by the second client, a second image recognition request carrying the second image and the key information, and identifying the second image to obtain a second image recognition copy;

Calculating the key information according to a preset rule to obtain a keyword key, and establishing a queue indexed by the key, and placing the first image recognition copy and the second image recognition copy into the queue for storing ;

When receiving the matching request of the first client and the second client, reading the first image recognition copy and the second image recognition copy according to the key to the queue;

Matching the first image recognition copy and the second image recognition copy to obtain a matching result between the first client and the second client.
The method of claim 12, wherein when the matching request of the first client and the second client is received, the method further includes:

The timing is started starting from the reception time of the matching request, and the time duration of the matching process is obtained.
The method according to claim 12 or 13, wherein before the reading of the first image recognition copy and the second image recognition copy in the queue, the method further comprises:

Determining whether the timing duration exceeds the rated duration of the current match;

If yes, the current matching operation is aborted;

If not, a read operation of each image recognition copy is performed.
The method according to any one of claims 12-14, wherein reading the first image recognition copy and the second image recognition copy into the queue comprises:

Counting the image recognition copy in the queue to obtain a counting result;

If the counting result is 2, performing a matching step of each image recognition scheme;

If the count result is less than 2, the image recognition copy in the queue is read again at the specified time interval.
The method according to any one of claims 12-15, wherein the interval specifying time to read the image recognition copy in the queue again comprises: reading the image recognition copy in the queue again at intervals of 1 ms.
The method according to any one of claims 12 to 16, wherein when the matching request of the first client and the second client is received, the first image is read into the queue according to a key Identifying the copy and the second image recognition copy, including:

Parsing the matching request to obtain the key information carried therein;

Calculating the key information according to the preset rule, and obtaining a key corresponding to the key information;

Matching the key corresponding to the key information with each queue in the database, and finding a matching queue indexed by the key;

Reading the first image recognition copy and the second image recognition copy into the queue.
The method according to any one of claims 12-17, wherein the first image recognition copy and the second image recognition copy are matched to obtain the first client and the second client Match results, including:

And performing weighting calculation on the first image recognition copy and the second image recognition copy respectively, and obtaining weighting values of the first image recognition copy and the second image recognition copy;

Comparing the weight values of the first image recognition copy and the second image recognition copy to obtain an image matching result, and the image matching result is used as a matching result between the first client and the second client.
The method according to any one of claims 12 to 18, wherein weighting calculation is performed on the first image recognition copy and the second image recognition copy, respectively, to obtain the first image recognition copy and the second The weighted value of the image recognition copy, including:

When the first image and/or the second image includes a plurality of main lines, a weight and a weight ratio are separately set for the copy information of each main line;

The image recognition copy is weighted according to the weight of each main line and the weight ratio to obtain the total weight value of each image recognition copy.
An image matching device includes:

a receiving module, configured to receive an image matching request from a client, where the image matching request carries key information of the client;

a calculation module, configured to calculate the key information according to a preset rule to obtain a first keyword key;

a queue matching module, configured to match a queue corresponding to the first key in a queue indexed by a different key stored in a database, wherein each queue stores images of different clients but having the same key Image recognition copy;

The image matching module is adapted to read the image recognition copy in the matched queue, and match each image recognition copy according to the preset matching logic to obtain an image matching result, and obtain the client and other customers according to the image matching result. The matching result of the end.
The device according to claim 20, further comprising:

The timing module is adapted to receive the image matching request from the client, start timing according to the receiving time, and obtain the timing duration.
The device according to claim 20 or 21, wherein

The timing module is further adapted to determine whether the timing duration exceeds a rated duration of the current match before reading the image recognition copy in the matched queue;

If yes, the current matching operation is aborted;

If not, the image matching module is triggered to perform a read operation of each image recognition copy.
The apparatus according to any one of claims 20 to 22, further comprising:

The sending module is configured to send an error code that is unsuccessful to the client after the suspension of the matching operation.
The apparatus according to any one of claims 20 to 23, wherein the image matching module is further adapted to:

Counting the image recognition copy in the matched queue to obtain a counting result;

If the counting result is greater than or equal to 2, performing a matching step of each image recognition scheme;

If the count result is less than 2, the image recognition copy in the matched queue is read again at the specified time interval.
The apparatus of any of claims 20-24, wherein the image matching module is further adapted to read the image recognition copy in the matched queue again at intervals of 1 ms.
The apparatus of any one of claims 20-25, wherein the image matching module is further adapted to:

Performing a weighting calculation on each image recognition copy to obtain a weighted value of each image recognition copy;

The weighted values of each image recognition copy are compared to obtain an image matching result.
The apparatus of any one of claims 20-26, wherein the image matching module is further adapted to:

When each image includes a plurality of main lines, weights and weight ratios are separately set for the copy information of each main line;

Each image recognition copy is weighted to obtain a total weighted value of each image recognition copy.
The apparatus of any one of claims 20-27, wherein the image matching module is further adapted to:

The queue also stores an uploading image to obtain a client ID of the image recognition copy, and stores the client ID corresponding to the image recognition copy of the uploaded image;

When the image matching result is obtained, the corresponding client ID is reversely searched according to each image recognition copy participating in the matching operation, thereby confirming the client corresponding to the client ID;

The image matching result is displayed as a match result between the client and other clients confirmed.
The apparatus according to any one of claims 20 to 28, further comprising:

The selection module is adapted to obtain a matching result between the client and the other client based on the image matching result, and select another client that has the highest matching degree with the client according to the matching result between the client and the other client; Another client and the corresponding matching result are returned to the client.
The apparatus according to any one of claims 20 to 29, further comprising:

The copy identification module is adapted to generate an image recognition copy in each queue as follows:

Receiving at least two images from a client, and the at least two images carry the same key information;

Performing image recognition on the at least two images to obtain an image recognition result;

Performing calculation on the key information of the current image according to the preset rule to obtain a corresponding key;

A queue indexed by the key is created in the database for the key, and each image recognition copy is placed in the queue.
An image matching device includes:

The first receiving module is configured to receive a first image recognition request from the first client that carries the first image and the key information, and identify the first image to obtain a first image recognition copy;

The second receiving module is configured to receive a second image recognition request from the second client that carries the second image and the key information, and identify the second image to obtain a second image recognition copy;

a storage module, configured to calculate the key information according to a preset rule, obtain a keyword key, and establish a queue indexed by the key, and insert the first image recognition copy and the second image recognition copy Stored in the queue;

a reading module, configured to read the first image recognition copy and the second image recognition copy into the queue when receiving a matching request between the first client and the second client;

The matching module is adapted to match the first image recognition copy and the second image recognition copy to obtain a matching result between the first client and the second client.
The device according to claim 31, further comprising:

The timing module is adapted to start timing when the matching request of the first client and the second client is received, and start timing according to the receiving moment of the matching request, and obtain a timing duration of the matching process.
The apparatus according to claim 31 or 32, wherein said timing module is further adapted to: before reading said first image recognition copy and said second image recognition copy in said queue,

Determining whether the timing duration exceeds the rated duration of the current match;

If yes, the current matching operation is aborted;

If not, a read operation of each image recognition copy is performed.
The apparatus according to any one of claims 31 to 33, wherein the reading module is further adapted to: count an image recognition copy in the queue to obtain a counting result;

If the counting result is 2, performing a matching step of each image recognition scheme;

If the count result is less than 2, the image recognition copy in the queue is read again at the specified time interval.
The apparatus of any of claims 31-34, wherein the reading module is further adapted to read the image recognition copy in the queue again at intervals of 1 ms.
The apparatus of any of claims 31-35, wherein the reading module is further adapted to:

Parsing the matching request to obtain the key information carried therein;

Calculating the key information according to the preset rule, and obtaining a key corresponding to the key information;

Matching the key corresponding to the key information with each queue in the database, and finding a matching queue indexed by the key;

Reading the first image recognition copy and the second image recognition copy into the queue.
The apparatus of any of claims 31-36, wherein the matching module is further adapted to:

And performing weighting calculation on the first image recognition copy and the second image recognition copy respectively, and obtaining weighting values of the first image recognition copy and the second image recognition copy;

Comparing the weight values of the first image recognition copy and the second image recognition copy to obtain an image matching result, and the image matching result is used as a matching result between the first client and the second client.
The apparatus of any of claims 31-37, wherein the matching module is further adapted to:

When the first image and/or the second image includes a plurality of main lines, a weight and a weight ratio are separately set for the copy information of each main line;

The image recognition copy is weighted according to the weight of each main line and the weight ratio to obtain the total weight value of each image recognition copy.
A computer program comprising computer readable code that, when executed on a computing device, causes the computing device to perform the image matching method of any of claims 1-19.
A computer readable medium storing the computer program of claim 39.