Character recognition in distorted images
First Claim
1. A method comprising:
- (a) prior to rendering an imitation image, estimating one or more rendering parameters from a distorted image;
(b) rendering, by a processor the imitation image, the imitation image comprising one or more simulated characters that correspond to one or more characters within the distorted image;
(c) applying one or more distortion models to the imitation image, thereby generating a distorted imitation image;
(d) comparing the distorted imitation image with the distorted image to compute a similarity between the distorted imitation image and the distorted image; and
(e) identifying the one or more characters as the one or more simulated characters based on the similarity.
1 Assignment
0 Petitions
Accused Products
Abstract
Systems and methods are provided for recognizing characters within a distorted image. According to a one aspect, a method for recognizing one or more characters within a distorted image includes rendering one or more imitation images, the imitation images including simulations of the distorted image, applying one or more distortion models to the imitation images, thereby generating distorted imitation images, comparing the distorted imitation images with the distorted image in order to compute similarities between the distorted imitation images and the distorted image, and identifying the characters based on the best similarity. According to other aspects, the systems and methods can be configured to provide recognition of other distorted data types and elements.
-
Citations
21 Claims
-
1. A method comprising:
-
(a) prior to rendering an imitation image, estimating one or more rendering parameters from a distorted image; (b) rendering, by a processor the imitation image, the imitation image comprising one or more simulated characters that correspond to one or more characters within the distorted image; (c) applying one or more distortion models to the imitation image, thereby generating a distorted imitation image; (d) comparing the distorted imitation image with the distorted image to compute a similarity between the distorted imitation image and the distorted image; and (e) identifying the one or more characters as the one or more simulated characters based on the similarity. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18)
-
-
19. A system comprising one or more processors configured to:
-
(a) render an imitation image, the imitation image comprising one or more simulated characters that correspond to one or more characters within a distorted image; (b) apply at least a first distortion model to the imitation image, thereby generating an initially distorted imitation image; (c) compare the initially distorted imitation image with the distorted image by i. simulating a further distortion of the initially distorted imitation image, the further distortion comprising at least a second distortion model, and ii. computing a similarity between the further distortion of the initially distorted imitation image and the distorted image, the similarity comprising a probability that the one or more simulated characters are present in the distorted image; and (d) identify the one or more characters as the one or more simulated characters based on the similarity.
-
-
20. A non-transitory computer-readable medium comprising instructions that, when executed by a processing device, cause the processing device to:
-
(a) receive a distorted data element; (b) render a plurality of imitation data elements, each of the imitation data elements comprising a unique undistorted simulation of the distorted data element; (c) determine one or more distortion models to be applied to each of the imitation data elements, the one or more distortion models being applied based on at least one of; i. metadata that corresponds to the distorted data element, ii. a processing of the distorted data element in order to determine one or more distortion types present in the distorted data element, or iii. a processing of one or more data elements related to the distorted data element in order to determine one or more distortion types present in the one or more data elements related to the distorted data element; (d) apply the one or more distortion models to each of the imitation data elements, thereby generating a plurality of distorted imitation data elements; (e) compare each of the distorted imitation data elements with the distorted data element in order to compute a similarity between each respective distorted imitation data element and the distorted data element; (f) compare each of the computed similarities with one another to identify a most accurate simulation; and (g) output at least one of;
each of the computed similarities or the most accurate simulation. - View Dependent Claims (21)
-
Specification