Back to results
US 12,394,229 B1Granted
System and method for processing images to be uploaded to a server
AssigneeCharles Schwab & Co., Inc.
InventorsDavid Okun, Amy L. Acuff, Charles E. Gotlieb
Life of this patent
Estimating the term…
Abstract
A system and method selects an image from multiple images taken at nearly the same time, based on blur, contour value or an estimate of either or both of these derived from the accelerometer of the device that captured the images, and then uploads the selected image.
Claims21
A method of uploading at least one of a plurality of images of a subject from a mobile device to a server, the method comprising:
- recording readings of acceleration from an accelerometer of the mobile device while capturing the plurality of images of the subject in succession in a burst using a camera of the mobile device;
- mapping the readings of the acceleration from the accelerometer with respective ones of the plurality of images of the subject by correlating timestamps associated with the plurality of images of the subject with timestamps associated with the readings of the acceleration from the accelerometer;
- narrowing which of the plurality of images to perform a blur analysis on by predicting which of the plurality of images of the subject are least likely blurred images or low contoured images after capturing the burst of the plurality of images based on the mapping of the readings of the acceleration recorded at the timestamps associated with the respective ones of the plurality of images of the subject by,
- determining which of the readings of acceleration from the accelerometer are a lowest reading of acceleration after capturing the burst of the plurality of images,
- determining the timestamp associated with the lowest reading of acceleration, and
- determining the least likely blurred images or the low contoured images as ones of the plurality of images having a same timestamp as the timestamp associated with the lowest reading of acceleration;
- checking only the least likely blurred images or the low contoured images for recognizability from among the plurality of images by performing the blur analysis on only the least likely blurred images or the low contoured images among the plurality of images to generate a blur value and comparing the blur value with a threshold selected based on a maximum amount of blur that can be reliably recognized via optical character recognition (OCR);
- and
- selectively uploading the least likely blurred images or the low contoured images of the subject to the server based on results of the checking such that only a subset of the plurality of images captured by the mobile device during the burst are uploaded to the server for the server to perform the OCR thereon.
6 more elements
A system for uploading at least one of a plurality of images of a subject from a mobile device to a server, comprising:
- a memory and a processor, the memory storing a computer readable code that, when executed by the processor, configures the system to,
- record readings of acceleration from an accelerometer of the mobile device while capturing the plurality of images of the subject in succession in a burst using a camera of the mobile device;
- map the readings of the acceleration from the accelerometer with respective ones of the plurality of images of the subject by correlating timestamps associated with the plurality of images of the subject with timestamps associated with the readings of the acceleration from the accelerometer;
- narrow which of the plurality of images to perform a blur analysis on by predicting which of the plurality of images of the subject are least likely blurred images or low contoured images based on the mapping of the readings of the acceleration recorded at the timestamps associated with the respective ones of the plurality of images of the subject by,
- determining which of the readings of acceleration from the accelerometer are a lowest reading of acceleration after capturing the burst of the plurality of images,
- determining the timestamp associated with the lowest reading of acceleration, and
- determining the least likely blurred images or the low contoured images as ones of the plurality of images having a same timestamp as the timestamp associated with the lowest reading of acceleration;
- check only the least likely blurred images or the low contoured images for recognizability from among the plurality of images by performing the blur analysis on only the least likely blurred images or the low contoured images among the plurality of images to generate a blur value and comparing the blur value with a threshold selected based on a maximum amount of blur that can be reliably recognized via optical character recognition (OCR);
- and
- selectively upload the least likely blurred images or the low contoured images of the subject to the server based on results of the checking such that only a subset of the plurality of images captured by the mobile device during the burst are uploaded to the server for the server to perform the OCR thereon.
7 more elements
A non-transitory computer readable medium having computer readable program code embodied therein for uploading at least one of a plurality of images of a subject from a mobile device to a server, the computer readable program code configured to cause a computer system to:
- record readings of acceleration from an accelerometer of the mobile device while capturing the plurality of images of the subject in succession in a burst using a camera of the mobile device;
- map the readings of the acceleration from the accelerometer with respective ones of the plurality of images of the subject by correlating timestamps associated with the plurality of images of the subject with timestamps associated with the readings of the acceleration from the accelerometer;
- narrow which of the plurality of images to perform a blur analysis on by predicting which of the plurality of images of the subject are least likely blurred images or low contoured images based on the mapping of the readings of the acceleration recorded at the timestamps associated with the respective ones of the plurality of images of the subject by,
- determining which of the readings of acceleration from the accelerometer are a lowest reading of acceleration after capturing the burst of the plurality of images,
- determining the timestamp associated with the lowest reading of acceleration, and
- determining the least likely blurred images or the low contoured images as ones of the plurality of images having a same timestamp as the timestamp associated with the lowest reading of acceleration;
- check only the least likely blurred images or the low contoured images for recognizability from among the plurality of images by performing the blur analysis on only the least likely blurred images or the low contoured images among the plurality of images to generate a blur value and comparing the blur value with a threshold selected based on a maximum amount of blur that can be reliably recognized via optical character recognition (OCR);
- and
- selectively upload the least likely blurred images or the low contoured images of the subject to the server based on results of the checking such that only a subset of the plurality of images captured by the mobile device during the burst are uploaded to the server for the server to perform the OCR thereon.
6 more elements
Dependencies are inferred from the claim text.