Document conversion correlation engine
View Patent ↗Embodiments of the disclosure provide a system for correlating document pages. The system receives a source image of a document page and a rendered image of a markup language page converted from the document page. The system then performs a correlation analysis between the source image and the rendered image. Next, the system determines a correlation factor between the source image and the rendered image based on the correlation analysis, wherein the correlation factor indicates a page fidelity between the document page and the markup language page converted from the document page.
1. A method of correlating document images, the method comprising:
receiving a source image representing an original document page;
receiving a rendered image representing a markup language page rendered by a browser, the markup language page generated by converting one or more elements in the original document page into markup language elements;
performing, by a computer, a correlation analysis comprising image comparison between the source image and the rendered image to determine a page fidelity of the converted markup language page;
determining, by the computer, a correlation factor based on the correlation analysis, the correlation factor indicating the page fidelity of the converted markup language page compared to the original document page, wherein the correlation factor is a number within a numeric range, the numeric range extending between a first numeric value signifying no correlation and a second numeric value signifying a perfect match; and
responsive to the correlation factor exceeding a specified threshold value, determining that the converted markup language page has passed a validation check.
2. The method of claim 1 , wherein the one or more elements in the original document page comprises at least one of an image, a graph, a table, a formula, and a body of texts.
3. The method of claim 1 , wherein converting one or more elements in the original document page into markup language elements further comprises capturing position and composition information of the one or more elements in the original document page to preserve page fidelity.
4. The method of claim 1 , wherein the rendered image of the markup language page is rendered by a specific web browser on a specific software platform.
5. The method of claim 1 , wherein the image comparison between the source image and the rendered image comprises applying at least one of:
a text comparison algorithm;
an image comparison algorithm;
a principal components analysis (PCA) algorithm; a linear discriminant analysis (LDA) algorithm; and
any combination of the above algorithms.
6. The method of claim 1 , wherein the numeric range extends between 0 indicating no correlation and 100 indicating a perfect match.
7. The method of claim 1 , wherein:
the specified threshold value of the correlation factor corresponds to a requirement of the page fidelity based on a type and layout of the original document page.
8. The method of claim 1 , further comprising:
responsive to the correlation factor being lower than the specified threshold value, determining that the converted markup language page is subject to further review.
9. A non-transitory computer-readable storage medium storing executable computer program instructions for document images correlation, the computer program instructions comprising instructions for:
receiving a source image representing an original document page;
receiving a rendered image representing a markup language page rendered by a browser, the markup language page generated by converting one or more elements in the original document page into markup language elements;
performing a correlation analysis comprising image comparison between the source image and the rendered image to determine a page fidelity of the converted markup language page;
determining a correlation factor based on the correlation analysis, the correlation factor indicating the page fidelity of the converted markup language page compared to the original document page, wherein the correlation factor is a number within a numeric range, the numeric range extending between a first numeric value signifying no correlation and a second numeric value signifying a perfect match; and
responsive to the correlation factor exceeding a specified threshold value, determining that the converted markup language page has passed a validation check.
10. The non-transitory computer-readable medium of claim 9 , wherein the one or more elements in the original document page comprises at least one of an image, a graph, a table, a formula, and a body of texts.
11. The non-transitory computer-readable medium of claim 9 , wherein converting one or more elements in the original document page into markup language elements further comprises capturing position and composition information of the one or more elements in the original document page to preserve page fidelity.
12. The non-transitory computer-readable medium of claim 9 , wherein the rendered image of the markup language page is rendered by a specific web browser on a specific software platform.
13. The non-transitory computer-readable medium of claim 9 , wherein the image comparison between the source image and the rendered image comprises applying at least one of:
a text comparison algorithm;
an image comparison algorithm; a principal components analysis (PCA) algorithm;
a linear discriminant analysis (LDA) algorithm; and
any combination of the above algorithms.
14. The non-transitory computer-readable medium of claim 9 , wherein the numeric range extends between 0 indicating no correlation and 100 indicating a perfect match.
15. The non-transitory computer-readable medium of claim 9 , wherein:
the specified threshold value of the correlation factor corresponds to a requirement of the page fidelity based on a type and layout of the original document page.
16. The non-transitory computer-readable medium of claim 9 , further comprising instructions for:
responsive to the correlation factor being lower than the specified threshold value, determining that the converted markup language page is subject to further review.
17. A system for correlating document images, the system comprises:
a processor;
a receiving mechanism configured to receive a source image representing an original document page, and a rendered image representing a markup language page rendered by a browser, the markup language page generated by converting one or more elements in the original document page into markup language elements;
a correlation-analysis mechanism, coupled to the processor, configured to perform a correlation analysis comprising image comparison between the source image and the rendered image to determine a page fidelity of the converted markup language page, determine a correlation factor based on the correlation analysis, the correlation factor indicating the page fidelity of the converted markup language page compared to the original document page, the correlation factor being a number within a numeric range, the numeric range extending between a first numeric value signifying no correlation and a second numeric value signifying a perfect match, and responsive to the correlation factor exceeding a specified threshold value, determine that the converted markup language page has passed a validation check.
18. The system of claim 17 , wherein the one or more elements in the original document page comprises at least one of an image, a graph, a table, a formula, and a body of texts.
19. The system of claim 17 , wherein converting one or more elements in the original document page into markup language elements further comprises capturing position and composition information of the one or more elements in the original document page to preserve page fidelity.
20. The system of claim 17 , wherein the rendered image of the markup language page is rendered by a specific web browser on a specific software platform.
21. The system of claim 17 , wherein the image comparison between the source image and the rendered image comprises applying at least one of:
a text comparison algorithm;
an image comparison algorithm; a principal components analysis (PCA) algorithm;
a linear discriminant analysis (LDA) algorithm; and
any combination of the above algorithms.
22. The system of claim 17 , wherein the numeric range extends between 0 indicating no correlation and 100 indicating a perfect match.
23. The system of claim 17 , wherein the specified threshold value of the correlation factor corresponds to a requirement of the page fidelity based on a type and layout of the original document page.
24. The system of claim 17 , wherein the correlation-analysis mechanism is further configured to:
responsive to the correlation factor being lower than the specified threshold value, determine that the converted markup language page is subject to further review.