Ancestry.com Operations Inc.

Shawn Reid

Jack Reese

Laryn Brown

Michael Murdock

Randy Hansen

Mark Oswald

Donald B Curtis

Systems and methods for classifying historical images. A feature extractor may create feature vectors corresponding to a plurality of images. A first classification of the plurality of images may be performed based on the plurality of feature vectors, which may include assigning a label to each of the plurality of images and assigning a probability for each of the assigned labels. The assigned probability for each of the assigned labels may be related to a statistical confidence that a particular assigned label is correctly assigned to a particular image. A subset of the plurality of images may be displayed to a display device. An input corresponding to replacement of an incorrect label with a corrected label for a certain image may be received from a user. A second classification of the plurality of images based on the input from the user may be performed.

Clustering historical images using a convolutional neural net and labeled data bootstrapping

Systems and methods for registering documents. A two-dimensional document image and one or more form images may be obtained. The document image may be projected onto a horizontal axis and a vertical axis to create a horizontal document projection and a vertical document projection. The form images may be projected onto the horizontal axis and the vertical axis to create a horizontal form projection and a vertical form projection. The horizontal document projection may be correlated with the horizontal form projection of the form images and the vertical document projection may be correlated with the vertical form projection of the form images. Correlation scores may be calculated based on the correlations.

2D multi-coordinate adaptive dynamics programming method for robust document registration

A handwriting recognition system converts word images on documents, such as document images of historical records, into computer searchable text. Word images (snippets) on the document are located, and have multiple word features identified. For each word image, a word feature vector is created representing multiple word features. Based on the similarity of word features (e.g., the distance between feature vectors), similar words are grouped together in clusters, and a centroid that has features most representative of words in the cluster is selected. A digitized text word is selected for each cluster based on review of a centroid in the cluster, and is assigned to all words in that cluster and is used as computer searchable text for those word images where they appear in documents. An analyst may review clusters to permit refinement of the parameters used for grouping words in clusters, including the adjustment of weights and other factors used for determining the distance between feature vectors.

System and method for transcribing handwritten records using word grouping with assigned centroids

System and method for transcribing historical records into digitized text

System and method for transcribing handwritten records using word groupings based on feature vectors

According to one embodiment, a method of linking a document image with indexed data is provided. The method may be performed by providing a document image, which is a digitized document having various information. Indexed data is also provided, which is a record that includes information extracted from the document image or a different document image. The process is further performed by identifying a feature of the indexed data and analyzing the document image to determine whether the feature is present within the information of the digitized document. The feature may be information or a characteristic defined by the information extracted from the document image or the different document image. If the feature is present within the information of the digitized document, a determination is made that the indexed data corresponds with the document image and the indexed data is linked with the document image.

Automated field position linking of indexed data to digital images

Various methods and devices involving reprographic copy stands are described. A reprographic copy stand may include a first support element configured to support a first cover of a book at a first plurality of angles. The reprographic copy stand may also include a second support element configured to support a second configured to support a second cover of the book at a second plurality of angles. The reprographic copy stand may include a platform configured to couple with the first support element at a first plurality of positions and the second support element at a second plurality of positions. The distance between the support elements may be variable according to the position of the first support selected from the first plurality of positions and the position of the second support selected from the second plurality of positions such that a plurality of book binding thicknesses are accommodated by the distance.

Multi-angle and multi-position reprographic copy stand

Devices, systems, and methods for facilitating the imaging of books are presented. A first transparent plate and a second transparent plate may be positioned at approximately a 45 degree angle to the first transparent plate. A dual-sided mirror may be configured to be positioned substantially parallel with the first transparent plate and to be positioned substantially parallel with the second transparent plate. The first side of the dual-sided mirror may be configured to reflect a first image of a first page of a book, the first image being received by the dual-sided mirror through the first transparent plate. The second side of the dual-sided mirror may be configured to reflect a second image of a second page of the book, the second image being received by the dual-sided mirror through the second transparent plate.

Dual sided mirror book imaging devices and methods

The present invention provides a method of correcting alignment of an image. The method includes identifying an image of a document, and detecting a plurality of lines within the image. The method also includes classifying at least a subset of the plurality of lines as either horizontal or vertical lines, and determining whether a variance exists within a subset of the plurality of lines. Furthermore, the method includes applying a corrective transformation to adjust the image, where the variance within the subset of the plurality of lines is at least partially reduced as compared to the variance prior to the application of the corrective transformation.

Multi-perpendicular line-based deskew with affine and perspective distortion correction

A method of processing an image includes receiving a digital version of the image, processing the digital version of the image through at least two binarization processes to thereby create a first binarization and a second binarization, and processing the first binarization through a first optical character recognition process to thereby create a first OCR output file. Processing the first binarization through a first optical character recognition process includes compiling first metrics associated with the first OCR output file. The method also includes processing the second binarization through the first optical character recognition process to thereby create a second OCR output file. Processing the second binarization through the first optical character recognition process includes compiling second metrics associated with the second OCR output file. The method also includes using the metrics, at least in part, to select a final OCR output file from among the OCR output files.

Growing community of inventors

Orem, UT, United States of America

Shawn Reid