# US 2019/0245692 A1 · Compression and manipulation-resistant fuzzy hashing

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

- Number: US 2019/0245692 A1
- Document ID: US20190245692A1
- Status: Published application
- Kind: A1
- Application number: 15/888,613
- Filed: 2018-02-05
- Published: 2019-08-08
- Earliest claimed: 2018-02-05
- Assignees: Colossio, Inc.
- Applicants: Colossio, Inc.
- Inventors: Joseph A. Jaroch
- Web page: https://patentgrep.com/patent/US20190245692A1

## CPC

- H04L 9/0643: {Hash functions, e.g. MD5, SHA, HMAC or f9 MAC}
- G06F 21/602: {Providing cryptographic facilities or services}
- G06F 21/64: Protecting data integrity, e.g. using checksums, certificates or signatures
- G06V 10/50: by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
- G06V 20/80: Recognising image objects characterised by unique random patterns
- H04L 9/3239: {involving non-keyed hash functions, e.g. modification detection codes [MDCs], MD5, SHA or RIPEMD}
- G06V 20/95: {Pattern authentication; Markers therefor; Forgery detection}
- H04L 2209/08: Randomization, e.g. dummy operations or using noise
- H04L 2209/34: Encoding or coding, e.g. Huffman coding or error correction
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

Systems and a method for compression and manipulation-resistant fuzzy hashing are provided. In one or more aspects, a system includes a network interface to receive an image object from a network, and a processor to process the image object. The processing includes generating pairs of random numbers using a hash of pixel data of the image object as a seed. The processing further includes identifying a number of coordinate pairs, within image pixels of the image object, such that coordinate values of each coordinate pair of the identified coordinate pairs approximately matches one pair of the random numbers. A number of first entropy values associated with first sub-areas corresponding to the identified coordinate pairs are determined. An anchor point within the image pixels is identified that has coordinate values corresponding to a sub-area that is associated with a highest entropy value among the determined first entropy values.

## Claims

### Claim 1

A system for generating a compression and manipulation-resistant fuzzy hash for an image object, the system comprising:
a network interface configured to receive an image object from a network; and
a processor configured to process the image object by:
generating a plurality of pairs of random numbers using a hash of pixel data associated with the image object as a seed;
identifying a plurality of coordinate pairs, within image pixels of the image object, wherein coordinate values of each coordinate pair of the identified plurality of coordinate pairs approximately matches one pair of random numbers of the plurality of pairs of random numbers;
determining a plurality of first entropy values associated with a plurality of first sub-areas of the image pixels of the image object corresponding to the identified plurality of coordinate pairs; and
identifying an anchor point within the image pixels of the image object, the identified anchor point having coordinate values corresponding to a sub-area of the plurality of first sub-areas of the image pixels that is associated with a highest entropy value among the determined plurality of first entropy values associated with the plurality of first sub-areas.

### Claim 2 (depends on 1)

The system of claim 1, wherein the processor is configured to store coordinate values of the identified anchor point, and wherein each sub-area of the plurality of first sub-areas comprise a circle with a radius value related to at least one dimension of the image object.

### Claim 3 (depends on 2)

The system of claim 2, wherein the processor is configured to determine a plurality of second entropy values corresponding to a plurality of second sub-areas of the image pixels of the image object.

### Claim 4 (depends on 3)

The system of claim 3, wherein the processor is further configured to determine the plurality of second sub-areas of the image pixels of the image object by first starting from the anchor point and moving forward to a last pixel of the image object and then moving forward from a first pixel of the image object to the anchor point.

### Claim 5 (depends on 4)

The system of claim 4, wherein moving forward is performed by incrementing coordinate values by a predetermined value, and wherein the predetermined value is dependent on dimensions of the image object.

### Claim 6 (depends on 4)

The system of claim 4, wherein the processor is further configured to determine standard deviation and mean values of the plurality of second entropy values and to store values of a sub-set of the plurality of second entropy values that are within the standard deviation value from the mean value.

### Claim 7 (depends on 6)

The system of claim 6, wherein the processor is further configured to:
rank the values of the sub-set of the plurality of second entropy values in one of an ascending or a descending order,
round the ranked values of the sub-set of the plurality of second entropy values to a predetermined precision, and
determine a checksum of the rounded, ranked values of the sub-set of the plurality of second entropy values.

### Claim 8 (depends on 7)

The system of claim 7, wherein the processor is further configured to compare the checksum against a plurality of stored checksums of a database to determine whether the image object is original or is created based on an image corresponding to a flagged checksum of the stored checksums.

### Claim 9 (depends on 1)

The system of claim 1, wherein the processor is further configured to determine the hash of pixel data associated with the image object using a secure algorithm (SHA) 256, and wherein the plurality of coordinate pairs comprise Cartesian coordinates pairs.

### Claim 10 (depends on 1)

The system of claim 1, wherein the processor is further configured to determine the plurality of first entropy values using a Shannon entropy calculation scheme.

### Claim 11 (depends on 1)

The system of claim 1, wherein the image object comprises an image or a frame of a video clip.

### Claim 12

A system comprising:
memory; and
a processor coupled to the memory and configured to receive an image from a network, the processor configured to:
identify a plurality of coordinate pairs, within image pixels of the image, based on a plurality of pairs of random numbers;
determine a plurality of first entropy values associated with a plurality of first sub-areas of the image pixels of the image, each sub-area of the plurality of first sub-areas corresponding to a coordinate pair of the identified plurality of coordinate pairs;
identify an anchor point, within the image pixels of the image, associated with a center point of a sub-area of the plurality of first sub-areas of the image pixels having a highest entropy value among the plurality of first entropy values associated with the plurality of first sub-areas.

### Claim 13 (depends on 12)

The system of claim 12, wherein the coordinate values of each coordinate pair of the identified plurality of coordinate pairs approximately matches one pair of random numbers of a plurality of pairs of random numbers generated using a hash of the image.

### Claim 14 (depends on 13)

The system of claim 13, wherein the processor is further configured to determine the hash of the image by using a hash of pixel data associated with the image as a seed.

### Claim 15 (depends on 13)

The system of claim 13, wherein the identified anchor point has coordinate values that correspond to the center point of the sub-area of the plurality of first sub-areas, and wherein each of the plurality of first sub-areas comprises a circle with a radius having a value equal to a fraction of a width of the image.

### Claim 16 (depends on 12)

The system of claim 12, wherein the processor is further configured to:
identify a plurality of second sub-areas of the image pixels of the image by scanning pixels of the image while incrementing coordinate values by a predetermined value that is related to a dimension of the image, and
determine a plurality of second entropy values corresponding to the plurality of second sub-areas of the image pixels of the image.

### Claim 17 (depends on 16)

The system of claim 16, wherein the processor is further configured to:
store values of a sub-set of the plurality of second entropy values that are within a standard deviation value from a mean value of the plurality of second entropy values, and
rank the values of the sub-set of the plurality of second entropy values in one of an ascending or a descending order.

### Claim 18 (depends on 17)

The system of claim 17, wherein the processor is further configured to determine a checksum of rounded, ranked values of the sub-set of the plurality of second entropy values, and wherein the rounded, ranked values of the sub-set of the plurality of second entropy values are generated by rounding the ranked values of the sub-set of the plurality of second entropy values to a 1/100 precision.

### Claim 19

A method of generating a compression and manipulation-resistant fuzzy hash for an image, the method comprising:
identifying a plurality of coordinate pairs, within image pixels of the image, wherein coordinate values of each coordinate pair of the identified plurality of coordinate pairs approximately matches one pair of random numbers of a plurality of pairs of random numbers;
determining a plurality of first entropy values associated with a plurality of first sub-areas of the image pixels of the image that correspond to the identified plurality of coordinate pairs; and
determining a highest entropy value among the plurality of first entropy values and a sub-area of the plurality of first sub-areas corresponding the highest entropy value; and
identifying an anchor point within the image pixels of the image, the identified anchor point having coordinate values matching coordinate values of the determined sub-area.

### Claim 20 (depends on 19)

The method of claim 19, further comprising forming an array of second entropy values corresponding to a group of second sub-areas, wherein sub-areas of the group of second sub-areas are obtained via scanning the image starting from the anchor point and ending in the anchor point, and further comprising ranking the array of second entropy values and rounding values of the ranked array of second entropy values and determining a checksum of the rounded values of the ranked array of second entropy values.
