Two-dimensional code identification and positioning
The present specification provides a two-dimensional code identification method and device, and a two-dimensional code positioning and identification model establishment method and device. The two-dimensional code identification method includes: obtaining a to-be-identified two-dimensional code, and performing global feature positioning detection on the to-be-identified two-dimensional code by using a pre-established two-dimensional code positioning and identification model; performing focus adjustment, based on a predetermined image resolution, on the to-be-identified two-dimensional code on which positioning detection is performed; and decoding the to-be-identified two-dimensional code on which focus adjustment is performed. The present specification can improve the identification accuracy of two-dimensional codes shot in complex scenarios.
1. A computer-implemented method, comprising:
obtaining an image of a reference two-dimensional code;
obtaining a first resolution of the image of the reference two-dimensional code;
obtaining a sample two-dimensional code based on a predetermined environment condition;
obtaining a second resolution of the sample two-dimensional code;
determining that the second resolution of the sample two-dimensional code matches the first resolution of the reference two-dimensional code;
positioning global features of the reference two-dimensional code based on the sample two-dimensional code and identifier information of the sample two-dimensional code, wherein the identifier information comprises one or more global features of the sample two-dimensional code;
providing the reference two-dimensional code and the global features of the reference two-dimensional code as training data to train a two-dimensional code positioning and identification model;
detecting a third two-dimensional code within a second obtained image by performing global feature positioning detection using the two-dimensional code positioning and identification model;
subsequent to detecting the third two-dimensional code, obtaining a focus-adjusted image of the third two-dimensional code based on a predetermined image resolution; and
decoding the third two-dimensional code from the focus-adjusted image of the third two-dimensional code.
2. The computer-implemented method of claim 1 , wherein the method comprises:
determining that a resolution of the second obtained image of the third two-dimensional code fails to satisfy the predetermined image resolution; and obtaining the focus-adjusted image of the third two-dimensional code based on the predetermined image resolution responsive to determining that the resolution of the third two-dimensional code fails to satisfy the predetermined image resolution.
3. The computer-implemented method of claim 1 , wherein obtaining the focus-adjusted image of the third two-dimensional code based on the predetermined image resolution comprises applying an auto-focusing algorithm to obtain an updated image of the third two-dimensional code at a different resolution.
4. The computer-implemented method of claim 1 , wherein obtaining the focus-adjusted image of the third two-dimensional code based on the predetermined image resolution comprises applying a zoom adjustment based on the predetermined image resolution to obtain an updated image of the third two-dimensional code at a different resolution.
5. The computer-implemented method of claim 4 , wherein applying the zoom adjustment comprises:
detecting whether a two-dimensional code scanning device comprises an optical zoom function; and
based on detecting that the two-dimensional code scanning device has the optical zoom function, adjusting the two-dimensional code scanning device to a new zoom level, and capturing a revised image of the third two-dimensional code at the new zoom level, or
based on detecting that the two-dimensional code scanning device does not have the optical zoom function, applying an image interpolation process to a captured pixel area of the second obtained image of the third two-dimensional code.
6. The computer-implemented method of claim 5 , wherein, after capturing the revised image of the third two-dimensional code, the method further comprises:
determining the third two-dimensional code does not satisfy the predetermined image resolution; and
applying an interpolation process to the captured pixel area of the revised image of the third two-dimensional code.
7. The computer-implemented method of claim 1 , wherein the two-dimensional code positioning and identification model comprises a machine learning network and the method comprises generating the machine learning network, wherein generating the machine learning network comprises:
labeling corresponding identifier information for a specified area of the sample two-dimensional code; and
performing training by using the sample two-dimensional code and the identifier information as input data for the machine learning network to obtain the two-dimensional code positioning and identification model.
8. The computer-implemented method of claim 7 , wherein the machine learning network is a convolution neural network, deep learning network, deep convolutional neural network, regions with convolutional neural networks (R-CNN), Faster R-CNN, regression-based detection methods, you only look once (YOLO), or single-shot detector (SSD).
9. The method according to claim 7 , wherein the specified area is a corner point of the sample two-dimensional code.
10. A non-transitory, computer-readable medium storing one or more instructions executable by a computer system to perform operations comprising:
obtaining an image of a reference two-dimensional code;
obtaining a first resolution of the image of the reference two-dimensional code;
obtaining a sample two-dimensional code based on a predetermined environment condition;
obtaining a second resolution of the sample two-dimensional code;
determining that the second resolution of the sample two-dimensional code matches the first resolution of the reference two-dimensional code;
positioning global features of the reference two-dimensional code based on the sample two-dimensional code and identifier information of the sample two-dimensional code, wherein the identifier information comprises one or more global features of the sample two-dimensional code;
providing the reference two-dimensional code and the global features of the reference two-dimensional code as training data to train a two-dimensional code positioning and identification model;
detecting a third two-dimensional code within a second obtained image by performing global feature positioning detection using the two-dimensional code positioning and identification model;
subsequent to detecting the third two-dimensional code, obtaining a focus-adjusted image of the third two-dimensional code based on a predetermined image resolution; and
decoding the third two-dimensional code from the focus-adjusted image of the third two-dimensional code.
11. The non-transitory, computer-readable medium of claim 10 , wherein the operations comprise:
determining that a resolution of the second obtained image of the third two-dimensional code fails to satisfy the predetermined image resolution; and obtaining the focus-adjusted image of the third two-dimensional code based on the predetermined image resolution responsive to determining that the resolution of the third two-dimensional code fails to satisfy the predetermined image resolution.
12. The non-transitory, computer-readable medium of claim 10 , wherein obtaining the focus-adjusted image of the third two-dimensional code based on the predetermined image resolution comprises applying an auto-focusing algorithm to obtain an updated image of the third two-dimensional code at a different resolution.
13. The non-transitory, computer-readable medium of claim 10 , wherein obtaining the focus-adjusted image of the third two-dimensional code based on the predetermined image resolution comprises applying a zoom adjustment based on the predetermined image resolution to obtain an updated image of the third two-dimensional code at a different resolution.
14. The non-transitory, computer-readable medium of claim 13 , wherein applying the zoom adjustment comprises:
detecting whether a two-dimensional code scanning device comprises an optical zoom function; and
based on detecting that the two-dimensional code scanning device has the optical zoom function, adjusting the two-dimensional code scanning device to a new zoom level, and capturing a revised image of the third two-dimensional code at the new zoom level, or
based on detecting that the two-dimensional code scanning device does not have the optical zoom function, applying an image interpolation process to a captured pixel area of the second obtained image of the third two-dimensional code.
15. The non-transitory, computer-readable medium of claim 14 , wherein, after capturing the revised image of the third two-dimensional code, the operations further comprise:
determining the third two-dimensional code does not satisfy the predetermined image resolution; and
applying an interpolation process to the captured pixel area of the revised image of the third two-dimensional code.
16. The non-transitory, computer-readable medium of claim 10 , wherein the two-dimensional code positioning and identification model comprises a machine learning network and operations comprise generating the machine learning network, wherein generating the machine learning network comprises:
labeling corresponding identifier information for a specified area of the sample two-dimensional code; and
performing training by using the sample two-dimensional code and the identifier information as input data for the machine learning network to obtain the two-dimensional code positioning and identification model.
17. The non-transitory, computer-readable medium of claim 16 , wherein the machine learning network is a form of convolution neural network, deep learning network, deep convolutional neural network, regions with convolutional neural networks (R-CNN), Faster R-CNN, regression-based detection methods, you only look once (YOLO), or single-shot detector (SSD).
18. The non-transitory, computer-readable medium of claim 16 , wherein the specified area is a corner point of the sample two-dimensional code.
19. A computer-implemented system, comprising:
one or more computers; and
one or more computer memory devices interoperably coupled with the one or more computers and having tangible, non-transitory, machine-readable media storing one or more instructions that, when executed by the one or more computers, perform one or more operations comprising:
obtaining an image of a reference two-dimensional code;
obtaining a first resolution of the image of the reference two-dimensional code;
obtaining a sample two-dimensional code based on a predetermined environment condition;
obtaining a second resolution of the sample two-dimensional code;
determining that the second resolution of the sample two-dimensional code matches the first resolution of the reference two-dimensional code;
positioning global features of the reference two-dimensional code based on the sample two-dimensional code and identifier information of the sample two-dimensional code, wherein the identifier information comprises one or more global features of the sample two-dimensional code;
providing the reference two-dimensional code and the global features of the reference two-dimensional code as training data to train a two-dimensional code positioning and identification model;
detecting a third two-dimensional code within a second obtained image by performing global feature positioning detection using the two-dimensional code positioning and identification model;
subsequent to detecting the third two-dimensional code, obtaining a focus-adjusted image of the third two-dimensional code based on a predetermined image resolution; and
decoding the third two-dimensional code from the focus-adjusted image of the third two-dimensional code.
20. The computer-implemented system of claim 19 , wherein the operations comprise:
determining that a resolution of the second obtained image of the third two-dimensional code fails to satisfy the predetermined image resolution; and obtaining the focus-adjusted image of the third two-dimensional code based on the predetermined image resolution responsive to determining that the resolution of the third two-dimensional code fails to satisfy the predetermined image resolution.
21. The computer-implemented system of claim 19 , wherein obtaining the focus-adjusted image of the third two-dimensional code based on the predetermined image resolution comprises applying an auto-focusing algorithm to obtain an updated image of the third two-dimensional code at a different resolution.
22. The computer-implemented system of claim 19 , wherein obtaining the focus-adjusted image of the third two-dimensional code based on the predetermined image resolution comprises applying a zoom adjustment based on the predetermined image resolution to obtain an updated image of the third two-dimensional code at a different resolution.
23. The computer-implemented system of claim 22 , wherein applying the zoom adjustment comprises:
detecting whether a two-dimensional code scanning device comprises an optical zoom function; and
based on detecting that the two-dimensional code scanning device has the optical zoom function, adjusting the two-dimensional code scanning device to a new zoom level, and capturing a revised image of the third two-dimensional code at the new zoom level, or
based on detecting that the two-dimensional code scanning device does not have the optical zoom function, applying an image interpolation process to a captured pixel area of the second obtained image of the third two-dimensional code.
24. The computer-implemented system of claim 23 , wherein, after capturing the revised image of the third two-dimensional code, the operations further comprise:
determining the third two-dimensional code does not satisfy the predetermined image resolution; and
applying an interpolation process to the captured pixel area of the revised image of the third two-dimensional code.
25. The computer-implemented system of claim 19 , wherein the two-dimensional code positioning and identification model comprises a machine learning network and the operations comprise generating the machine learning network, wherein generating the machine learning network comprises:
labeling corresponding identifier information for a specified area of the sample two-dimensional code; and
performing training by using the sample two-dimensional code and the identifier information as input data for the machine learning network to obtain the two-dimensional code positioning and identification model.
26. The computer-implemented system of claim 25 , wherein the machine learning network is a form of convolution neural network, deep learning network, deep convolutional neural network, regions with convolutional neural networks (R-CNN), Faster R-CNN, regression-based detection methods, you only look once (YOLO), or single-shot detector (SSD).
27. The computer-implemented system of claim 25 , wherein the specified area is a corner point of the sample two-dimensional code.