Systems and methods for performing OCR of a series of images depicting text symbols. An example method comprises: receiving a current image of a series of images of an original document, wherein the current image at least partially overlaps with a previous image of the series of images; performing optical symbol recognition (OCR) of the current image to produce an OCR text and a corresponding text layout; identifying, using the OCR text and the corresponding text layout, a plurality of textual artifacts in each of the current image and the previous image, wherein each textual artifact is represented by a sequence of symbols that has a frequency of occurrence within the OCR text falling below a threshold frequency; identifying, in each of the current image and the previous image, a corresponding plurality of base points, wherein each base point is associated with at least one textural artifact of the plurality of textual artifacts; identifying, using coordinates of matching base points in the current image and the previous image, parameters of a coordinate transformation converting coordinates of the previous image into coordinates of the current image; associating, using the coordinate transformation, at least part of the OCR text with a cluster of a plurality of clusters of symbol sequences, wherein the OCR text is produced by processing the current image and wherein the symbol sequences are produced by processing one or more previously received images of the series of images; identifying, for each cluster, a median string representing the cluster of symbol sequences; and producing, using the median string, a resulting OCR text representing at least a portion of the original document.

Provided are a method and an apparatus for recognizing a tire by using an image of a tire captured by using a terminal. The apparatus recognizes a tire by converting an image of a round tire included in the image into a linear image, extracting an area of a character to be recognized from the linear image, and determining a character for learning, which is most similar to the extracted area of the character to be recognized from among a pre-constructed group of characters for learning, as a character in the area of the character to be recognized.

A method includes passing an original text document through distortion filter generators to generate a training dataset that includes distorted text documents. Each distortion filter generator is configured to distort words or letters of words in phrases of text of a facsimile image in a respective unique manner. A neural network model is trained to recognize each respective distortion and match each respective distortion with each respective distortion filter generator based on the training dataset and the original text document. Image data of one facsimile having at least one text distortion is received and inputted to the trained neural network model. The output of the trained neural network model is coupled to an input of an optical character recognition (OCR) engine. The trained neural network model and the OCR engine convert the received image data of the incoming facsimile corrected for the at least one text distortion to machine-encoded text.

A system uses a neural network based model to perform scene text recognition. The system achieves high accuracy of prediction of text from scenes based on a neural network architecture that uses double attention mechanism. The neural network based model includes a convolutional neural network component that outputs a set of visual features and an attention extractor neural network component that determines attention scores based on the visual features. The visual features and the attention scores are combined to generate mixed features that are provided as input to a character recognizer component that determines a second attention score and recognizes the characters based on the second attention score. The system trains the neural network based model by adjusting the neural network parameters to minimize a multi-class gradient harmonizing mechanism (GHM) loss. The multi-class GHM loss varies based on a level of difficulty of the sample.

A method, apparatus, device and storage medium for recognizing a bill image may include: performing text detection on a bill image, and determining an attribute information set and a relationship information set of each text box of at least two text boxes in the bill image; determining a type of the text box and an associated text box that has a structural relationship with the text box based on the attribute information set and the relationship information set of the text box; and extracting structured bill data of the bill image, based on the type of the text box and the associated text box that has the structural relationship with the text box.

An example non-transitory computer-readable medium includes instructions executable by a processor to detect boundaries of a representation of a document page in a captured image, model the boundaries of the representation of the document page as nonlinear curves, use the nonlinear curves to transform pixels of the representation of the document page into pixels of a dewarped representation of the document page, and output a dewarped image based on the dewarped representation of the document page.

An image processing method, an image processing device, an electronic device, and a storage medium are provided. The image processing method includes: obtaining an input image, wherein the input image includes M character rows; performing global correction processing on the input image to obtain an intermediate corrected image; determining the M character row lower boundaries; determining the relative offset of all pixels in the intermediate corrected image according to the M character row lower boundaries, the first image boundary and the second image boundary of the intermediate corrected image; determining the local adjustment offset of all pixels in the intermediate corrected image according to the relative offsets of all pixels in the intermediate corrected image; and performing local adjustment on the intermediate corrected image according to the local adjustment offsets of all pixels in the intermediate corrected image to obtain the target corrected image.

Systems and methods provide a deployable cloud-agnostic redaction container for performing optical character recognition and redacting information from a document using a cloud-based, guided redaction framework. An example method for document redaction includes receiving a plurality of documents and extracting pages from the plurality of documents. The method then determines, based on a load balancing criterion, a processing order for the pages extracted from the plurality of documents, and performs, based on the processing order, an optical character recognition process and a redaction process on the pages to generate redacted pages. The redacted pages are provided for transmission or storage to a cloud data management platform.

A license plate recognition system comprises an image capturing unit, for capturing an image; a license plate recognition unit, coupled to the image capture unit, for detecting a location of a license plate image in the image, correcting the license plate image according to at least one first corner of the license plate image, to generate a corrected license plate image, and recognizing the corrected license plate image, to generate a license plate recognition result; and an output unit, coupled to the license plate recognition unit, for outputting the license plate recognition result.

A text recognition method and apparatus that relate to the field of information processing technologies are provided. This effectively resolves a low recognition rate of curved text. The text recognition method includes: obtaining a to-be-detected image; determining a target text detection area in the to-be-detected image, where the target text detection area includes target text in the to-be-detected image, and the target text detection area is a polygonal area including m (m is a positive integer greater than 2) vertex pairs; correcting the polygonal area to m−1 rectangular areas to obtain a corrected target text detection area; and performing text recognition on the corrected target text detection area, and outputting the target text.