Interactively predicting fields in a form
Methods, apparatuses, and embodiments related to interactively predicting fields in a form. A computer system received an image of a form. A user moves a cursor to a first field of the form, and the computer system automatically displays a predicted location of the field, including a bounding box that represents a boundary of the field. The computer system further predicts the field name/label based on text in the document. The user clicks on the field to indicate that he wants to digitize the field. When needed, the user interactively modifies the size of the bounding box that represents the extent of the field, changes the name/label of the field. Once finalized, the user can cause the field information (e.g., the bounding box coordinate, the bounding box location, the name/label of the field, etc.) to be written to a database.
1. A method for predicting a location of a field on a form based on an image of the form, the method comprising:
receiving, by a computer system, a first image data of the form;
creating a grid data structure that distributes pixels from the first image data into blocks;
identifying, by the computer system, a plurality of visible line segments of the first image data by executing a computer vision analysis within each block of the grid;
predicting the location and a boundary of the field of the form using a first block as identified by a particular location of the image indicated via user input; and
a first line segment of the first block; and
determining coordinates of a rectangular box that represents the field based on the location and the boundary of the field based on determining the extent of the field based on other nearby line segments that indicate the extent of the field; and
displaying the boundary of the field at the location of the field.
2. The method of claim 1 , wherein each of the blocks is associated with a 3×3 grid of pixels.
3. The method of claim 1 , further comprising:
determining to digitize the field based on an indication by a user to create the field; and
digitizing the field by:
writing coordinates that define a boundary of the field to a database that is associated with the image; and
determining a label for the field by:
executing an optical character recognition (OCR) algorithm on a portion of the image that is determined based on the coordinates of the rectangular box that represents the field.
4. The method of claim 3 , wherein determining the label for the field further includes:
receiving selection of an autocomplete in response to the user typing a character to indicate the label for the field.
5. The method of claim 3 , further comprising:
receiving a second image of a filled-out version of the form;
locating an image region on the second image from coordinates corresponding to coordinates of the rectangular box on the first image data;
executing the optical character reading algorithm on the image region on the second image; and
determining a text value of the image region on the second image based on said executing the optical character reading algorithm on the portion of the second image.
6. The method of claim 1 , wherein said determining coordinates of the rectangular box includes determining that there are no other nearby line segments that indicate an extent of the field, wherein said determining the extent of the field includes determining an average or median height of text of the form, and wherein one dimension of the extent of the field is based on results of said determining the average or the median height of the text.
7. A method comprising:
receiving, by a computer system, a first image data of a form;
executing an image analysis algorithm, by the computer system that identifies visual characteristics based on the first image data, and that analyzes the visual data to determine which of the visual data represent visual line segments and text characters;
receiving an indication of a particular location on the first image data;
identifying a nearest visual line segment of the visual data associated with the particular location;
generating a first boundary of a fillable field on the form based on the nearest visual line segment;
identifying a set of text characters that are associated with the nearest visual line segment, the set of text characters including a text height;
generating a second boundary of the fillable field based on the text height;
forming the fillable field based on the first boundary and the second boundary;
determining a predicted label of the field based on the set of text characters;
in response to receipt of user input configured to create the field, writing data that indicates the fillable field and the predicted label to a database that is associated with the image;
receiving a second image of a filled-out version of the form;
locating an image region on the second image from coordinates corresponding to coordinates of the fillable field on the first image data;
executing the optical character reading algorithm on the image region on the second image; and
determining a text value of the image region on the second image based on said executing the optical character reading algorithm on the portion of the second image.
8. The method of claim 7 ,
wherein said receiving the indication of the particular location on the first image data is data from a mouse that controls a cursor.
9. The method of claim 7 , wherein the particular location of the image is a location of a pixel of the image.
10. The method of claim 7 , wherein the image region on the second image is based on use of oversized coordinates as corresponding to coordinates of the fillable field on the first image data.
11. The method of claim 7 , wherein the cursor indicating the particular location of the image includes the cursor being located within a predetermined distance of the particular location.
12. The method of claim 7 ,
wherein the field is one of a plurality of fields of a table,
wherein a subset of the visual line segments indicate the plurality of fields of the table, and
wherein the rectangular portion of the image includes the plurality of fields of the table.
13. The method of claim 12 , further comprising:
predicting a location and boundary of each of the plurality of fields of the table; and
displaying the boundary of said each of the plurality of fields.
14. A computing system comprising:
a processor;
a user input interface configured to receive an indication of a particular location on image data of a form; and
a memory coupled to the processor and storing instructions which, when executed by the processor, cause the computing system to perform operations including:
form a fillable field on the image data of the form through use of an image analysis algorithm that identifies visual characteristics of the image data including visual line segments and text characters, wherein the instructions further cause the processor to establish a set of coordinates of a rectangular box of the fillable field based on identification of a nearest visual line segment of associated with the particular location and that there are no other nearby line segments that indicate an extent of the field, and a determination of the extent of the field based on an average or median height of text characters of the form, and wherein one dimension of the extent of the field is based on results of said average or the median height of the text characters.
15. The computing system of claim 14 , wherein the user input interface is a peripheral mouse.
16. The computing system of claim 15 , wherein the processor further has instructions to:
determine a predicted label of the field based on the set of text characters.
17. The computing system of claim 14 , further comprising:
a display, wherein said formation of the fillable field includes displaying the fillable field on the display.
18. The computing system of claim 14 , wherein user input interface is further configured to receive text characters and deliver to the system for insertion into the fillable field.