US-20260129031-A1 - VIDEO AND STILL IMAGE DATA ALTERATION TO ENHANCE PRIVACY

US20260129031A1US 20260129031 A1US20260129031 A1US 20260129031A1US-20260129031-A1

Abstract

A computer alters at least one recognizable metric or text in a digitally encoded photographic image by operating an alteration algorithm in response to user input data while preserving an overall aesthetic quality of the image and obscuring an identity of at least one individual or geographic location appearing in the image. An altered digitally-encoded photographic image prepared by the altering of the at least one recognizable metric or text in the image is stored in a computer memory. User feedback and/or automatic analysis may be performed to define parameter values of the alteration algorithm such that the alteration process achieves preservation of aesthetic qualities while obscuring an identity of interest.

Inventors

Gary S. Shuster

Assignees

Gary S. Shuster

Dates

Publication Date: 20260507
Application Date: 20251229

Claims (19)

1 . A method, comprising: altering by a computer at least one portion of metadata associated with a digitally-encoded photographic image in conjunction with altering the digitally-encoded photographic image; preserving an aesthetic quality of the digitally-encoded photographic image while decreasing an accuracy of automated recognition of an identity of at least one individual appearing in the digitally-encoded photographic image, wherein the altering comprises altering selected facial recognition metrics; and storing, in a computer memory, the altered digitally-encoded photographic image in conjunction with the altered at least one portion of metadata.
2 . The method of claim 1 , further comprising decreasing the accuracy of automatic image recognition to below a defined minimum confidence level using an alteration algorithm.
3 . The method of claim 2 , further comprising modifying the altered selected facial recognition metrics to match, at least in part, at least one user-set parameter.
4 . The method of claim 2 , further comprising identifying a recognition parameter set based on an image recognition algorithm, wherein the recognition parameter set is defined by a set of parameter range values outside of which the image recognition algorithm cannot recognize the identity above the defined minimum confidence level.
5 . The method of claim 2 , further comprising modifying the alteration in response to the at least one user-set input specifying a desired confidence level.
6 . The method of claim 5 , further comprising defining parameter values for the alteration algorithm used in the altering of the image within the recognition parameter set.
7 . The method of claim 5 , further comprising controlling at least one parameter of the alteration algorithm so as to preserve the aesthetic quality of the image as determined by an aesthetic quality measuring algorithm to above a defined minimum preservation level.
8 . A method, comprising: altering by a computer at least one portion of metadata associated with a digitally-encoded photographic image in conjunction with altering the digitally-encoded photographic image; preserving an aesthetic quality of the digitally-encoded photographic image while decreasing an accuracy of automated recognition of an identity of at least one individual appearing in the digitally-encoded photographic image, wherein the altering comprises altering selected object recognition metrics; and storing, in a computer memory, the altered digitally-encoded photographic image in conjunction with the altered at least one portion of metadata.
9 . The method of claim 8 , further comprising decreasing the accuracy of automatic image recognition to below a defined minimum confidence level using an alteration algorithm.
10 . The method of claim 9 , further comprising modifying the altered selected facial recognition metrics to match, at least in part, at least one user-set parameter.
11 . The method of claim 9 , further comprising identifying a recognition parameter set based on an image recognition algorithm, wherein the recognition parameter set is defined by a set of parameter range values outside of which the image recognition algorithm cannot recognize the identity above the defined minimum confidence level.
12 . The method of claim 9 , further comprising defining parameter values for the alteration algorithm used in the altering of the image within the recognition parameter set.
13 . The method of claim 12 , further comprising controlling at least one parameter of the alteration algorithm so as to preserve the aesthetic quality of the image as determined by an aesthetic quality measuring algorithm to above a defined minimum preservation level.
14 . The method of claim 12 , further comprising modifying the portions of the object altered to match, at least in part, at least one user-set parameter.
15 . A method, comprising: training a model heuristically to recognize an aesthetic quality of an image to above a minimum aesthetic quality level, where the model is trained at least in part by user feedback over a sample set of images.
16 . The method of claim 15 , where the model adopts parameter values to achieve a designated amount of the aesthetic quality of the image.
17 . The method of claim 16 , where the designated amount of the aesthetic quality of the image is achieved by limiting an amount of image modification.
18 . The method of claim 17 , where the image modification is achieved in a manner that reduces an accuracy of at least one image recognition system.
19 . The method of claim 17 , where an aesthetic preservation parameter is set based on the designated amount of the aesthetic quality of an image, and a third set of parameters is defined as parameter values that are members of an aesthetic preservation parameter set and are not members of a recognition parameter set.

Description

CROSS-REFERENCE TO RELATED APPLICATION This application claims priority to and is a continuation of U.S. patent application Ser. No. 18/118,035, filed Mar. 6, 2023, which is a continuation of U.S. patent application Ser. No. 17/392,109, filed Aug. 2, 2021, (U.S. Pat. No. 11,600,108, issued Mar. 7, 2023), which is a continuation of U.S. patent application Ser. No. 16/292,216, filed Mar. 4, 2019, (U.S. Pat. No. 11,080,513, issued Aug. 3, 2021), which is a continuation-in-part of U.S. patent application Ser. No. 15/665,216, filed Jul. 31, 2017 (U.S. Pat. No. 10,223,576, issued Mar. 5, 2019), which is a continuation of U.S. patent application Ser. No. 14/853,782, filed Sep. 14, 2015 (U.S. Pat. No. 9,721,144, issued Aug. 1, 2017), which is a continuation of U.S. patent application Ser. No. 14/294,047, filed Jun. 2, 2014 (U.S. Pat. No. 9,135,493, issued Sep. 15, 2015), which is a continuation of U.S. patent application Ser. No. 13/349,546, filed Jan. 12, 2012 (U.S. Pat. No. 8,744,119, issued Jun. 3, 2014), which claims priority to U.S. Provisional Patent Application Ser. No. 61/431,965, filed Jan. 12, 2011, all of which are hereby incorporated by reference in their entireties. FIELD OF INVENTION This application relates to methods and systems for image data processing and image sharing over a network. DESCRIPTION OF THE BACKGROUND Anonymity on the internet is being destroyed by technology that allows searching databases of images for people. There are services such as “tineye” that find duplicate photographs, and technology such as Google's Picasa are capable of going much further, identifying a person in a photograph and then finding them in other photographs. For example, a Congressman's swearing-in photo might be used to develop the set of facial characteristics that are then looked for in other pictures, leading to the location of otherwise obscurely located embarrassing bachelor party pictures. See also the United States patent on the subject, “Method and Apparatus for Photograph Finding”, U.S. Pat. No. 7,454,498 and related continuations. It would be desirable for people who publish personal images on computer networks to conveniently ensure that posted images will not be subject to undesired image recognition applications, but solutions to provide such protection while still enabling online sharing of personal image data are either currently unavailable or suffer from substantial limitations in convenience or effectiveness. In addition to recognizing elements of still images, automated analysis of moving images can be used to identify a person via techniques such as gait recognition. One advantage of using body shape, characteristics and/or gait as a basis for identification of a person in a moving image is that the resolution necessary for recognition is reduced, allowing identification of persons at a far greater distance, with inferior focus, or in the background of a video. SUMMARY OF THE INVENTION Methods and systems for image processing to protect specified personal identities or geographic locations from automatic recognition, or graphic data alteration to enhance online privacy, are described in detailed in the detailed description, and certain aspects are described or summarized below. This summary and the following detailed description should be interpreted as complementary parts of an integrated disclosure, which parts may include redundant subject matter and/or supplemental subject matter. Omissions in either section do not indicate priority or relative importance of any element described in the integrated application. Differences between the sections may include supplemental disclosures of alternative embodiments or additional details, or alternative descriptions of identical embodiments using different terminology, depending on the nature of the differences. In an aspect, a method for graphic data alteration to enhance online privacy may include a number of steps, all of which may be performed by a computer, either automatically or semi-automatically. The method may include altering at least one recognizable metric or text in a digitally-encoded photographic image by a computer operating an alteration algorithm in response to user input data while preserving an overall aesthetic quality of the image and obscuring an identity of at least one individual or geographic location appearing in the image. The method may further include storing, in a computer memory, an altered digitally-encoded photographic image prepared by the altering of the at least one recognizable metric or text in the digitally-encoded photographic image. Altering the image may include altering selected facial recognition metrics. In addition, or in the alternative, altering the image may include scrambling text. As used herein, an “aesthetic quality” means a quality determined by the subjective taste of an individual human being, or collectively by an aggregate of subjective tastes of a group of individual human beings. An ae