# US 11,908,172 B2 · Methods and systems to reduce privacy invasion and methods and systems to thwart same

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 11,908,172 B2
- Document ID: US11908172B2
- Status: Granted
- Kind: B2
- Application number: 17/383,218
- Filed: 2021-07-22
- Published: 2022-02-24
- Granted: 2024-02-20
- Earliest claimed: 2020-08-19
- Assignees: Privacy4Cars, Inc.
- Applicants: Privacy4Cars, LLC
- Inventors: Andrea Amico
- Web page: https://patentgrep.com/patent/US11908172B2

## CPC

- G06V 10/443: {by matching or filtering}
- G06V 30/10: Character recognition
- G06F 18/211: Selection of the most significant subset of features
- G06F 18/22: Matching criteria, e.g. proximity measures
- G06V 10/42: Global feature extraction by analysis of the whole pattern, e.g. using frequency domain transformations or autocorrelation
- G06V 10/56: relating to colour
- G06V 40/10: Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V 20/625: {License plates}
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Robust recognition systems to identify potentially identifying information that is contained within a dataset representing a system target are disclosed, together with related method. In an implementations, the systems and methods receive the dataset, identify features in the dataset that have a characteristic indicative of an expected feature, processes the identified features to yield a candidate feature, wherein the candidate feature is one feature contained within the identified features that has the highest probability of containing potentially identifying information.

## Claims

### Claim 1

A method for a recognition system to identify potentially identifying information that is contained within a dataset representing a system target comprising:
receiving the dataset;
identifying features in the received dataset that have a characteristic indicative of an expected feature, wherein one or more features within the identified features are decoy features having a decoy characteristic similar to the characteristic indicative of an expected feature;
the decoy feature configured to mimic one or more expected features of the system target; and
processing the identified features to yield a candidate feature, wherein the candidate feature is one feature contained among the identified features that has the highest probability of containing potentially identifying information, the candidate feature distinct from the one or more decoy features.

### Claim 2 (depends on 1)

A method for a recognition system as set forth in claim 1, wherein the characteristic indicative of an expected feature is selected from the group consisting of one or more of the following: (i) whether the size of the feature is at generally the same size or a greater size than an expected features, (ii) whether the geometry of the features is generally consistent with the expected features, and (iii) whether the color of the features is generally consistent with the expected features.

### Claim 3 (depends on 1)

A method for a recognition system as set forth in claim 1, wherein the processing step comprises:
removing the one or more decoy features within the identified features.

### Claim 4 (depends on 3)

A method for a recognition system as set forth in claim 3, wherein the removing step comprises:
processing the features within the identified features to identify any information contained therein and processing any such information to reveal whether such information might be potentially identifying information.

### Claim 5 (depends on 4)

A method for a recognition system as set forth in claim 4, wherein the processing step includes the step of comparing the identified information with information in a pre-populated dataset that is pre-populated by the system.

### Claim 6 (depends on 5)

A method for a recognition system as set forth in claim 5, wherein the pre-populated dataset contains information that is flagged by the system such that the system identifies the feature associated with the information as the candidate feature if such information of a feature matches information in the pre-populated dataset.

### Claim 7 (depends on 5)

A method for a recognition system as set forth in claim 5, wherein the pre-populated dataset contains information that is known by the system as decoy information such that the system removes the feature associated with the information from the set of identified features if such information of a features matches information in the pre-populated dataset.

### Claim 8 (depends on 1)

A method as set forth in claim 1, wherein the system target includes a smokescreen that is included in the identified features.

### Claim 9 (depends on 1)

A method as set forth in claim 1, wherein the potentially identifying information is a license plate number and the features expected by the recognition system are associated with the shape of a license plate having the license plate number.

### Claim 10 (depends on 1)

A method as set forth in claim 1, wherein the potentially identifying information is one or more persons, the features expected by the recognition system are associated with one or more tattoos.