System and method for determining compression rates for images comprising text
AssigneeGlassbox Ltd.
InventorsEliav ATOUN, Oleg BEZRUDNY
Life of this patent
Estimating the term…
Abstract
A system for determining compression rates for images, the system comprising a processing resource configured to: obtain a given image at least partially comprising a given text; compress the given image at a given compression ratio, giving rise to a compressed image; perform Optical Character Recognition (OCR) on the compressed image, giving rise to OCR text; compare the OCR text to the given text, giving rise to comparison results; upon the comparison results meeting a rule, increase the given compression rate; and upon the compression results not meeting a rule, return to a previous compression rate, if any.
Claims20
A system for determining compression rates for images, the system comprising a processing resource configured to:
- provide for each screen of one or more screens of an application:
- (a) text displayed on the respective screen, (b) a compression ratio associated with the respective screen, if any, and (c) a compression ratio determination status associated with the respective screen;
- receive, from a given device executing the application, a compressed screenshot of a given screen of the screens of the application, compressed in accordance with the compression ratio associated with the given screen, or, upon no compression ratio being associated with the given screen, in accordance with a default compression ratio;
- and
- upon the provided compression ratio determination status of the given screen being indicative of a non-final compression ratio, perform the following:
- perform Optical Character Recognition (OCR), using OCR software, on the compressed screenshot, giving rise to OCR text;
- determine an accuracy score by comparing the OCR text to the text provided for the given screen;
- upon the accuracy score being below a threshold, set the compression ratio determination status of the given screen to be indicative of a final compression ratio;
- and
- upon the accuracy score being equal to or above a threshold, increase the compression ratio.
A system for determining compression rates for images, the system comprising a processing resource configured to:
- obtain a given image at least partially comprising a given text;
- compress the given image at a given compression ratio, giving rise to a compressed image;
- perform Optical Character Recognition (OCR) on the compressed image, giving rise to OCR text;
- compare the OCR text to the given text, giving rise to comparison results;
- upon the comparison results meeting a rule, increase the given compression rate;
- and
- upon the compression results not meeting a rule, return to a previous compression rate, if any.
A method for determining compression rates for images, the method comprising:
- providing, by a processing resource, for each screen of one or more screens of an application:
- (a) text displayed on the respective screen, (b) a compression ratio associated with the respective screen, if any, and (c) a compression ratio determination status associated with the respective screen;
- receiving, by the processing resource, from a given device executing the application, a compressed screenshot of a given screen of the screens of the application, compressed in accordance with the compression ratio associated with the given screen, or, upon no compression ratio being associated with the given screen, in accordance with a default compression ratio;
- and
- upon the provided compression ratio determination status of the given screen being indicative of a non-final compression ratio, performing the following by the processing resource:
- perform Optical Character Recognition (OCR), using OCR software, on the compressed screenshot, giving rise to OCR text;
- determine an accuracy score by comparing the OCR text to the text provided for the given screen;
- upon the accuracy score being below a threshold, set the compression ratio determination status of the given screen to be indicative of a final compression ratio;
- and
- upon the accuracy score being equal to or above a threshold, increase the compression ratio.
A method for determining compression rates for images, the method comprising:
- obtaining, by a processing resource, a given image at least partially comprising a given text;
- compressing, by the processing resource, the given image at a given compression ratio, giving rise to a compressed image;
- performing, by the processing resource, Optical Character Recognition (OCR) on the compressed image, giving rise to OCR text;
- comparing, by the processing resource, the OCR text to the given text, giving rise to comparison results;
- upon the comparison results meeting a rule, increasing, by the processing resource, the given compression rate;
- and
- upon the compression results not meeting a rule, returning, by the processing resource, to a previous compression rate, if any.
A non-transitory computer readable storage medium having computer readable program code embodied therewith, the computer readable program code, executable by at least one processing resource of a computer to perform a method comprising:
- obtaining, by the processing resource, a given image at least partially comprising a given text;
- compressing, by the processing resource, the given image at a given compression ratio, giving rise to a compressed image;
- performing, by the processing resource, Optical Character Recognition (OCR) on the compressed image, giving rise to OCR text;
- comparing, by the processing resource, the OCR text to the given text, giving rise to comparison results;
- upon the comparison results meeting a rule, increasing, by the processing resource, the given compression rate;
- and
- upon the compression results not meeting a rule, returning, by the processing resource, to a previous compression rate, if any.
A non-transitory computer readable storage medium having computer readable program code embodied therewith, the computer readable program code, executable by at least one processing resource of a computer to perform a method comprising:
- providing, by a processing resource, for each screen of one or more screens of an application:
- (a) text displayed on the respective screen, (b) a compression ratio associated with the respective screen, if any, and (c) a compression ratio determination status associated with the respective screen;
- receiving, by the processing resource, from a given device executing the application, a compressed screenshot of a given screen of the screens of the application, compressed in accordance with the compression ratio associated with the given screen, or, upon no compression ratio being associated with the given screen, in accordance with a default compression ratio;
- and
- upon the provided compression ratio determination status of the given screen being indicative of a non-final compression ratio, performing the following by the processing resource:
- perform Optical Character Recognition (OCR), using OCR software, on the compressed screenshot, giving rise to OCR text;
- determine an accuracy score by comparing the OCR text to the text provided for the given screen;
- upon the accuracy score being below a threshold, set the compression ratio determination status of the given screen to be indicative of a final compression ratio;
- and
- upon the accuracy score being equal to or above a threshold, increase the compression ratio.
Dependencies are inferred from the claim text.