WO2021056812A1

WO2021056812A1 - Graphic adaptation system and graphic adaptation method based on same

Info

Publication number: WO2021056812A1
Application number: PCT/CN2019/122009
Authority: WO
Inventors: 刘树东; 陆鹏; 朱向升; 李�赫
Original assignee: 同程网络科技股份有限公司
Priority date: 2019-09-28
Filing date: 2019-11-29
Publication date: 2021-04-01
Also published as: CN110765396A; DE202020105321U1

Abstract

Disclosed in the present application are a graphic adaptation system and a graphic adaptation method based on same. The system comprises: a Web server, configured to check whether a picture uploaded by a user is a duplicate, and only upload a picture that is not a duplicate; a write server, configured to obtain and store the picture uploaded by the Web server; a tag information reading module, configured to receive and analyze a transmitted picture URL to obtain a corresponding picture id and a picture cutting style; a meta-information reading module, configured to obtain meta-information of a corresponding picture and original picture file information according to the picture id; and a picture cutting module, configured to cut, by using the obtained picture cutting style, an original picture obtained by the meta-information reading module.

Description

Graphic adaptation system and graphic adaptation method based on the system

This disclosure claims the priority of a Chinese patent application filed with the Chinese Patent Office with an application number of 201910928402.0 on September 28, 2019, and the entire content of the above application is incorporated into this disclosure by reference.

Technical field

This application relates to the technical field of graphics adaptation, for example, to a graphics adaptation system and a graphics adaptation method based on the system.

Background technique

With the development of the Internet, pictures often appear in large numbers in the business scenarios of major Internet companies due to their intuitive and vivid advantages, such as review pictures on APPs, hotel previews, etc. At the same time, since pictures need to be adapted to web sites and different mobile terminals, a lot of copies of the same picture need to be stored to adapt to different terminals.

For example, the Chinese patent with publication number CN101699839A discloses a mobile terminal graphics adaptation method, which includes the following steps: the user terminal sends a data packet containing the mobile phone number, the type of the mobile terminal and the specified picture to the WEB or WAP server; WEB Or WAP server calculates the terminal screen aspect ratio according to the mobile terminal type, and then queries the picture database server for pictures matching the terminal screen aspect ratio and picture number; WEB or WAP server sends the adapted picture to the mobile with the number terminal.

However, the above-mentioned related technical solutions have the following defects: there are multiple storage of the same picture, which requires a lot of storage costs. Therefore, in order to adapt to different terminals, while reducing storage capacity and saving costs as much as possible, it is necessary to design a In the case of storing only one picture, the picture can be cut in real time according to the terminal type, and the solution for real-time feedback of pictures of different sizes to the business side.

Summary of the invention

This application provides a graphics adaptation system and a graphics adaptation method based on the system.

This application provides a graphics adaptation system, which can save a lot of storage costs;

The present application provides a graphics adaptation method, which can reduce the pressure on the server while being able to feed back a picture of a corresponding size to the business party in real time.

The graphics adaptation system of this application is realized through the following technical solutions:

A graphics adaptation system, including:

The web server is set to query whether the pictures uploaded by the user are duplicated, upload the pictures that are not duplicated, and directly return the corresponding picture id for the duplicate pictures;

The writing server is configured to obtain and save the pictures uploaded by the Web server, and is also configured to manage and maintain meta-information of the saved pictures, the meta-information including the picture id;

The tag information reading module is set to receive and parse the transmitted picture URL corresponding to the user terminal type to obtain the corresponding picture id and picture cut style;

The meta-information reading module is configured to obtain the meta-information of the corresponding picture and the original image file information from the writing server according to the picture id obtained by the tag information reading module;

The image cutting module is configured to use the image cutting style acquired by the tag information reading module to cut the original image acquired by the meta information reading module and return it to the user terminal.

By adopting the above technical solution and adopting the architecture design of separation of reading and writing, in the case of storing only one picture, the picture can be cut in real time according to the user terminal type (APP, browser webpage, etc.), and different feedbacks are given to the business side in real time. Pictures of different sizes can be adapted to the user terminal, saving a lot of storage costs, and avoiding the situation that multiple sizes of the same picture are stored due to different sizes.

Optionally, the Web server includes:

The deduplication module is set to process the picture uploaded by the user into a picture in which all pixels are only black and white, and the processed picture is scaled down to form a contrast image, which is also set as the contrast image to be formed Compare with the comparison images corresponding to all the pictures in the writing server and judge whether they are duplicates;

The storage module is configured to store the comparison images of all the pictures in the writing server, and is also configured to store the comparison images obtained by the deduplication module when the deduplication module judges no.

By adopting the above-mentioned technical solution, deduplication is performed when users upload pictures, which saves the storage capacity of the system and avoids repeated pictures uploading.

Optionally, it further includes a preprocessing module configured to preprocess the picture uploaded by the Web server into a picture with a set format and a set style specification and upload it to the writing server.

By adopting the above technical solution, by default, real-time image cutting is performed for the uploaded original image with pixels larger than the set value, and the unified processing is performed to the set style specification, which limits the image storage size from the source, thereby saving system storage Capacity, reduce system load pressure.

Optionally, the tag information reading module includes:

The parsing unit is set to parse the transmitted picture URL corresponding to the user terminal to obtain the tag information tag and picture id of the picture cut;

The judging unit is set to judge whether the image needs to be cut according to the tag information tag;

The reading unit is set to construct the original image tag of the picture and upload it to the meta-information reading module when the judgment unit judges no; it is also set to read when the judgment unit judges yes Uploading the cut-out style in the tag information tag to the meta-information reading module;

Wherein, the tag information tag includes first feature information used to indicate whether the image is cut and second feature information including the image cut style.

By adopting the above technical solution, after obtaining the original image information and the crop size style, the original image can be cropped using the crop style. The cutting is flexible and configurable. By configuring the specifications and styles, the requirements of different sizes can be met. There is no need to traverse the pictures each time to generate pictures of different sizes, which reduces the pressure on the system. At the same time, various parameters such as whether to zoom or not can be configured during the cutting process to meet the requirements of different sizes and improve the adaptability of the system.

The graphics adaptation method of this application is realized through the following technical solutions:

A graphic adaptation method, which is based on the graphic adaptation system described in the above technical solution, and includes a picture upload method A and a picture cut method B;

The picture uploading method A includes the following steps:

S100. Image verification and de-duplication: De-duplication is performed when the user uploads the image, and the image id is directly returned to the duplicate image, and no repeated upload is required;

S200. Forwarding and backup: forwarding the pictures judged to be non-repetitive in step S100 to the writing server and writing the files, while managing and maintaining the meta information of the pictures, the meta information of each picture includes the picture id;

The picture cutting method B includes the following steps:

S300. Obtain the picture cut specification: parse the transmitted picture URL corresponding to the user terminal type and obtain the picture id and picture cut style;

S400. Obtaining image meta information: obtaining the meta information of the image and the file information of the original image according to the image id;

S500. Cut the picture: After obtaining the meta information of the picture, the file information of the original picture and the picture cut style, use the picture cut style to cut the original picture and return it to the user terminal.

By adopting the above technical solutions, the picture system saves only one copy of the original picture, cooperates with the style rules, and according to the style needs of the user, each request dynamically cuts the picture, and returns pictures of different sizes, which can reduce the pressure on the server while being able to real-time Feedback the effect of the corresponding size picture for the business side.

Optionally, step S100 includes the following sub-steps:

S101. Process the picture uploaded by the user through the MD5 information digest algorithm and generate a hash value;

S102. Compare the hash value generated in sub-step S101 with the hash values of all pictures in the writing server and determine whether it is repeated; if it is repeated, go to sub-step S103; if not, go to sub-step S104;

S103: Directly return the corresponding picture id for the repeated pictures, and do not repeatedly upload them;

S104: Store the hash value generated in sub-step S101, and then go to step S200.

By adopting the above technical solution, the user-uploaded picture can be quickly compared with the existing picture through the hash value, thereby achieving the purpose of deduplication and avoiding repeated uploading and repeated storage. On the other hand, even if the picture uploaded by the user is slightly different from the picture stored in the system, it can be distinguished by the hash value when the picture cannot be clearly distinguished by the naked eye, so as to avoid repeated uploads. Finally, because the hash value generated by the processing is small, it is conducive to storage, thereby reducing the pressure on the system.

Optionally, the picture uploading method A further includes the following steps:

S600. Picture preprocessing: preprocessing the pictures judged to be non-repetitive in step S100 into pictures with fixed rules;

The step S600 includes the following sub-steps:

S601: Acquire basic information of pictures that are determined to be non-repetitive in step S100, where the basic information includes format and style specifications;

S602. Determine the format of the picture obtained in sub-step S601; if the picture is not in jpg format, go to sub-step S603; if the picture is in jpg format, go to sub-step S604;

S603: Convert the picture into a jpg format, and enter sub-step S604;

S604: Crop the picture into a picture with a set style and specification, and go to step S200.

By adopting the above technical solutions, the storage size of the pictures is limited from the source, which not only saves storage resources, but also facilitates the dynamic cropping of the stored original pictures by the system, and improves the response efficiency of the system.

Optionally, the step S300 includes the following sub-steps:

S301: Generate a picture URL according to the type of the terminal;

S302: parse the picture URL generated in sub-step S301 and obtain the picture id and the tag information tag of the picture cut;

S303. Determine whether the tag information tag of the picture cut obtained in sub-step S302 is the original image; if the judgment is yes, then go to sub-step S304; if the judgment is no, then go to sub-step S305;

S304, construct the original picture label of the picture, and enter step S400;

S305: Read the cut pattern in the tag information tag, and enter step S400.

By adopting the above technical solution, it is possible to quickly determine whether the image needs to be cut according to the label information tag. If the image does not need to be cut, the original image tag is directly output to read the original image information, and the tag information tag is read only if the image needs to be cut. The cutting pattern in the middle, for subsequent cutting. Through the above judgment method, the response efficiency of the system is improved, and the load pressure of the system is reduced.

Optionally, the picture URL is composed of a domain name, a tag information tag, and a picture id, and the tag information tag includes first feature information indicating whether to cut a picture and second feature information including a cut picture style.

By adopting the above technical solution, it is possible to quickly determine the image cutting information based on the tag information tag, and through the analysis of the image URL, it can also accurately correspond to the corresponding original image to perform dynamic operations on it.

In summary, this application has the following advantages:

1. Through the setting of Web server, writing server, tag information reading module, meta information reading module and picture cutting module, it has the effect of saving a lot of storage costs;

2. Through the settings of the deduplication module, storage module and preprocessing module, from the source, whether to store pictures and the size of the stored pictures is restricted, which greatly reduces the load pressure of the system;

3. Through the setting of the analysis unit, the judgment unit and the reading unit, it is conducive to quickly identify the user's style needs, so as to feed back pictures of different sizes to the user in real time.

Description of the drawings

FIG. 1 is a schematic diagram of the overall structure of a graphics adaptation system shown in Embodiment 1 of the present invention;

2 is a schematic structural diagram of another graphics adaptation system shown in Embodiment 1 of the present invention;

FIG. 3 is a flowchart of a graphics adaptation method shown in Embodiment 2 of the present invention;

4 is a flowchart of step S100 shown in the second embodiment of the present invention;

FIG. 5 is a flowchart of step S600 shown in the second embodiment of the present invention;

Fig. 6 is a flowchart of step S300 shown in the second embodiment of the present invention.

In the figure, 1. Web server; 11. Deduplication module; 12. Storage module; 2. Write server; 3. Tag information reading module; 31. Analysis unit; 32. Judgment unit; 33. Reading unit; 4. Meta-information reading module; 5. Picture cutting module; 6. Preprocessing module; 7. Uploading subsystem; 8. Picture cutting subsystem.

detailed description

The application will be further described in detail below in conjunction with the drawings.

Example one

1, a graphics adaptation system disclosed in this application includes an uploading subsystem 7 and a picture cutting subsystem 8. The uploading subsystem 7 includes a Web server 1, a writing server 2 and a preprocessing module 6, and a picture cutting subsystem 8 includes the tag information reading module 3, the meta information reading module 4 and the image cutting module 5.

The web server 1 is used for querying whether the pictures uploaded by the user are duplicated, uploading pictures that are not duplicated, and directly returning the corresponding picture id for the duplicated pictures, so as to bind the picture with the information input by the user.

The writing server 2 is used to obtain and save the pictures uploaded by the Web server 1, and it is also used to manage and maintain the meta-information of the saved pictures. In some embodiments, the meta-information of the pictures includes picture id, picture content information, picture classification information, etc. .

The tag information reading module 3 is used to receive and parse the transmitted picture URL corresponding to the user terminal type, and is used to obtain the corresponding picture id and picture cut style according to the parsed data.

The meta-information reading module 4 is used to obtain the meta-information of the corresponding picture and the original image file information from the writing server 2 according to the picture id obtained by the tag information reading module 3.

The image cutting module 5 is configured to use the image cutting style obtained by the tag information reading module 3 to crop the original image obtained by the meta-information reading module 4, and return the cropped image to the user terminal.

The preprocessing module 6 is used to preprocess the pictures uploaded by the Web server 1 into pictures with a set format and a set style specification, and then upload them to the writing server 2. In some embodiments, when the picture uploaded by the Web server 1 is not in the jpg format, the preprocessing module 6 converts the picture into the jpg format; when the pixels of the picture uploaded by the Web server 1 are greater than the set value, such as 1500, the picture is processed as 1500*1500 style specifications.

2, the Web server 1 includes a deduplication module 11 and a storage module 12. The deduplication module 11 is used for pixel processing of the picture uploaded by the user, so that all pixels of the picture have only two colors of black and white, that is, the color photo is processed into a black and white photo; the deduplication module 11 processes the color photo into a black and white photo and then restores it. The black and white photos will be scaled down to form a contrast image. In this embodiment, the deduplication module 11 will scale down the processed black and white pictures to a quarter of the original; in addition, the deduplication module 11 is also used to reduce The obtained comparison image is compared with the comparison images corresponding to all the pictures in the writing server 2, and it is judged whether it is a duplicate. The storage module 12 is used to store the comparison images of all the pictures in the writing server 2, and is also used to store the comparison images obtained by the deduplication module 11 when the deduplication module 11 determines that the pictures are not repeated.

2, the tag information reading module 3 includes an analysis unit 31, a judgment unit 32, and a reading unit 33. The parsing unit 31 is used to parse the image URL corresponding to the user terminal, so as to obtain the tag information tag and the image id of the image cut. In some embodiments, the image URL is derived from the request of the user terminal after the system parses the image. The domain name, the rules for cutting pictures, and the picture id composition. The judging unit 32 is used for judging whether the image needs to be cut according to the tag information tag, and the reading unit 33 is used for constructing the original image tag of the picture and uploading it to the meta-information reading module 4 when the judging unit 32 judges no, and the judging unit 32 judges If yes, the reading unit 33 reads the cut pattern in the tag information tag and uploads it to the meta-information reading module 4. It should be noted that the tag information tag includes first feature information used to indicate whether to cut the image and second feature information including the image cut style.

The implementation principle of the foregoing embodiment is as follows:

When a user uploads a picture, in order to save storage capacity and avoid repeated uploading of the picture, the Web server 1 first checks the picture and removes duplicates. For repeated pictures, the corresponding picture id is directly returned and corresponds to the information requested by the user. Repeat upload. After uploading unique pictures, the preprocessing module 6 preprocesses the uploaded pictures, that is, by default, real-time cutting is performed on pictures with pixels larger than the set value in the original uploaded picture, for example, if the pixels of the uploaded original picture are larger than 1500 When cutting the pictures in real time, it will be uniformly processed to the style specification of 1500*1500 and the picture format will be adjusted to jpg, thus limiting the picture storage size and format. Finally, forwarding and backup are performed, that is, the preprocessing module 6 forwards the processed pictures to the writing server 2 and writes them into the file. At the same time, the writing server 2 can manage and maintain the meta-information of the pictures.

When cutting the picture, the system saves only one original picture, cooperates with the style rules, and according to the style requirements of the user, every time a dynamic picture cutting is requested, pictures of different sizes can be returned. In some embodiments, after the user terminal sends the request, the tag information reading module 3 first parses the picture URL, and obtains the picture cut specifications according to the composition rule of the picture URL. Subsequently, the meta-information reading module 4 obtains the original information of the image and the file information of the original image according to the image id. Finally, after the image cutting module 5 obtains the original image information and the image cutting style, it uses the image cutting style to cut the original image. At the same time, various parameters such as whether to zoom or not can be configured during the cutting process to meet the requirements of different sizes. After the cropping is completed, the system will return the corresponding size picture to the requester.

Example two

Referring to FIG. 3, a graphics adaptation method based on the graphics adaptation system in the first embodiment disclosed in this application includes a picture upload method A and a picture cut method B.

Picture upload method A includes the following steps:

S100. Image verification and deduplication: deduplication is performed when the user uploads the image, and the image id is directly returned to the duplicate image, and the upload is not repeated.

S600. Picture preprocessing: preprocess the pictures judged to be non-repetitive in step S100 as original pictures, and then enter S200.

S200. Forwarding and backup: forwarding the picture preprocessed in step S600 to the writing server 2 and writing the file, while managing and maintaining the meta-information of the picture. The meta-information of each picture includes a picture id.

Picture cutting method B includes the following steps:

S300. Obtain the picture cut specification: parse the transmitted picture URL corresponding to the user terminal type and obtain the picture id and picture cut style. In some embodiments, the picture URL is composed of a domain name, a tag information tag, and a picture id. The tag information tag includes first feature information indicating whether to cut the picture and second feature information including the cut picture style.

S400. Obtain the picture meta information: Obtain the meta information of the picture and the file information of the original picture according to the picture id.

Referring to FIG. 4, step S100 includes the following sub-steps:

S101: Process the picture uploaded by the user through the MD5 information digest algorithm and generate a hash value;

S102. Compare the hash value generated in sub-step S101 with the hash values of all pictures in the writing server 2 and determine whether it is repeated; if it is repeated, go to sub-step S103; if not, go to sub-step S104;

Referring to FIG. 5, step S600 includes the following sub-steps:

S601: Acquire basic information of pictures that are determined to be non-repetitive in step S100, where the basic information includes the format of the picture and the style specification of the picture;

S602. Determine whether the format of the picture obtained in sub-step S601 is jpg; if the picture is not in jpg format, go to sub-step S603; if the picture is in jpg format, go to sub-step S604;

S603: Convert the picture into a jpg format, and enter sub-step S604;

S604: Crop the picture into an original picture with a set style and specification, and go to step S200.

Referring to FIG. 6, step S300 includes the following sub-steps:

S301: The terminal application generates a picture URL according to the type of the terminal;

S303. Determine whether the tag information tag of the picture cut obtained in sub-step S302 is the original image; if the determination is yes, then go to sub-step S304; if the judgment is no, then go to sub-step S305;

S304, construct the original picture label of the picture, and enter step S400;

S305: Read the cut pattern in the tag information tag, and enter step S400.

Claims

A graphics adaptation system, including:

The web server (1) is set to query whether the pictures uploaded by the user are duplicated, upload the pictures that are not duplicated, and directly return the corresponding picture id for the duplicate pictures;

The writing server (2) is set to obtain and save the pictures uploaded by the Web server (1), and it is also set to manage and maintain the meta information of the saved pictures, and the meta information includes the picture id;

The tag information reading module (3) is set to receive and parse the transmitted picture URL corresponding to the user terminal type to obtain the corresponding picture id and picture cut style;

The meta-information reading module (4) is configured to obtain the meta-information of the corresponding picture and the original picture file information from the writing server (2) according to the picture id obtained by the tag information reading module (3);

The image cutting module (5) is configured to use the image cutting style acquired by the tag information reading module (3) to cut the original image acquired by the meta information reading module (4) and return it to the user terminal.
The graphics adaptation system according to claim 1, wherein the Web server (1) comprises:

The deduplication module (11) is set to process the picture uploaded by the user into a picture in which all pixels are only black and white, and the processed picture is scaled down to form a contrast image. It is also set to form a Compare the comparison image of with the comparison images corresponding to all the pictures in the writing server (2) and judge whether they are duplicates;

The storage module (12) is configured to store the comparison images of all pictures in the writing server (2), and is also configured to, when the deduplication module (11) determines no, the deduplication module ( 11) The obtained comparison image is stored.
The graphics adaptation system according to claim 1, further comprising a preprocessing module (6), the preprocessing module (6) is configured to preprocess the pictures uploaded by the Web server (1) into After setting the format and setting style specifications, the pictures are uploaded to the writing server (2).
The graphics adaptation system according to claim 1, wherein the tag information reading module (3) comprises:

The parsing unit (31) is set to parse the transmitted picture URL corresponding to the user terminal to obtain the tag information tag and picture id of the picture cut;

The judging unit (32) is set to judge whether the image needs to be cut according to the tag information tag;

The reading unit (33) is set to construct the original picture tag and upload it to the meta-information reading module (4) when the judgment unit (32) judges no; it is also set to When the judgment unit (32) judges as yes, read the cut pattern in the tag information tag and upload it to the meta-information reading module (4);

Wherein, the tag information tag includes first feature information used to indicate whether the image is cut and second feature information including the image cut style.
A graphics adaptation method, which is based on the graphics adaptation system of claim 1, and includes a picture upload method A and a picture cut method B;

The picture uploading method A includes the following steps:

S100. Image verification and de-duplication: De-duplication is performed when the user uploads the image, and the image id is directly returned to the duplicate image, and no repeated upload is required;

S200. Forwarding and backup: forwarding the pictures judged to be non-repetitive in step S100 to the writing server (2) and writing them into files, while managing and maintaining the meta information of the pictures, the meta information of each picture includes a picture id;

The picture cutting method B includes the following steps:

S300. Obtain the picture cut specification: parse the transmitted picture URL corresponding to the user terminal type and obtain the picture id and picture cut style;

S400. Obtaining image meta information: obtaining the meta information of the image and the file information of the original image according to the image id;

S500. Cut the picture: After obtaining the meta information of the picture, the file information of the original picture and the picture cut style, use the picture cut style to cut the original picture and return it to the user terminal.
The graphics adaptation method according to claim 5, wherein step S100 comprises the following sub-steps:

S101: Process the picture uploaded by the user through the MD5 information digest algorithm and generate a hash value;

S102. Compare the hash value generated in sub-step S101 with the hash values of all pictures in the writing server (2) and determine whether it is repeated; if it is repeated, go to sub-step S103; if not, go to sub-step S104 ；

S103: Directly return the corresponding picture id for the repeated pictures, and do not repeatedly upload them;

S104: Store the hash value generated in sub-step S101, and then go to step S200.
The graphic adaptation method according to claim 5, wherein the picture uploading method A further comprises the following steps:

S600. Picture preprocessing: preprocessing the pictures judged to be non-repetitive in step S100 into pictures with fixed rules;

The step S600 includes the following sub-steps:

S601: Acquire basic information of pictures that are determined to be non-repetitive in step S100, where the basic information includes format and style specifications;

S602. Determine the format of the picture obtained in sub-step S601; if the picture is not in jpg format, go to sub-step S603; if the picture is in jpg format, go to sub-step S604;

S603: Convert the picture into a jpg format, and enter sub-step S604;

S604: Crop the picture into a picture with a set style and specification, and go to step S200.
The graphics adaptation method according to claim 5, wherein said step S300 comprises the following sub-steps:

S301: Generate a picture URL according to the type of the terminal;

S302: parse the picture URL generated in sub-step S301 and obtain the picture id and the tag information tag of the picture cut;

S303. Determine whether the tag information tag of the picture cut obtained in sub-step S302 is the original image; if the judgment is yes, then go to sub-step S304; if the judgment is no, then go to sub-step S305;

S304, construct the original picture label of the picture, and enter step S400;

S305: Read the cut pattern in the tag information tag, and enter step S400.
The graphic adaptation method according to claim 8, wherein the picture URL is composed of a domain name, a tag information tag, and a picture id, and the tag information tag includes first feature information for indicating whether to cut the picture and includes the cut picture The second feature information of the style.