# US 2025/0030717 A1 · Detecting polymorphic botnets using an image recognition platform

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

- Number: US 2025/0030717 A1
- Document ID: US20250030717A1
- Status: Published application
- Kind: A1
- Application number: 18/224,902
- Filed: 2023-07-21
- Published: 2025-01-23
- Earliest claimed: 2023-07-21
- Assignees: Bank of America Corporation
- Applicants: Bank of America Corporation
- Inventors: Elvis Nyamwange; Erik Dahl; Brian Jacobson; Pratap Dande; Hari Vuppala; Rahul Yaksh; Rahul Phadnis; Amer Ali; Sailesh Vezzu
- Web page: https://patentgrep.com/patent/US20250030717A1

## CPC

- H04L 63/1433: {Vulnerability analysis}
- H04L 63/1416: {Event detection, e.g. attack signature detection}
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)

## Abstract

Arrangements for detecting polymorphic botnets using a pattern matching engine are provided. The platform may train a pattern matching engine to generate image pattern statuses based on a training set of historical image patterns. The platform may automatically generate real-time interaction information based on monitoring interactions between user devices and cloud platforms. The platform may generate a potentially malicious image pattern based on the real-time interaction information. The platform may then input the image pattern into the pattern matching engine to generate an image pattern status for the image pattern. The platform may store the image pattern to a pattern storage repository. Based on identifying the image pattern as malicious, the platform may initiate a security action. The platform may update the pattern matching engine based on the image pattern and its image pattern status to detect future polymorphic botnet attacks.

## Claims

### Claim 1

A computing platform comprising:
at least one processor;
a communication interface communicatively coupled to the at least one processor; and
memory storing computer-readable instructions that, when executed by the at least one processor, cause the computing platform to:
train, based on a training set of historical image patterns stored in at least one pattern storage repository, a pattern matching engine to generate image pattern statuses, wherein the training set comprises a plurality of historical malicious image patterns and a plurality of historical safe image patterns, wherein a given historical image pattern comprises historical interaction information for a cloud-based network;
automatically generate real-time interaction information based on monitoring interactions between a first user device and a cloud platform;
generate a first potentially malicious image pattern based on the real-time interaction information;
input the first potentially malicious image pattern into the pattern matching engine to generate an image pattern status for the first potentially malicious image pattern, wherein generating the image pattern status comprises comparing the first potentially malicious image pattern to at least one historical image pattern;
identify, based on the comparing, that the first potentially malicious image pattern is a malicious image pattern, and in response:
store the first potentially malicious image pattern at a malicious pattern storage repository; and
initiate at least one security action responsive to malicious image patterns; and
update, using the first potentially malicious image pattern and the image pattern status for the first potentially malicious image pattern, the pattern matching engine, wherein updating the pattern matching engine causes the pattern matching engine to determine one or more additional image pattern statuses based on the image pattern status for the first potentially malicious image pattern.

### Claim 2 (depends on 1)

The computing platform of claim 1, wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, cause the computing platform to:
retrieve at least one signature of a known botnet from a cybersecurity platform;
input the at least one signature into a pattern creation engine to generate at least one known malicious image pattern; and
store the at least one known malicious image pattern at the at least one pattern storage repository.

### Claim 3 (depends on 1)

The computing platform of claim 1, wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, cause the computing platform to:
generate a second potentially malicious image pattern based on the real-time interaction information;
determine an image pattern status for the second potentially malicious image pattern based on inputting the second potentially malicious image pattern into the pattern matching engine;
store, based on a determination that the second potentially malicious image pattern is a safe image pattern, the second potentially malicious image pattern at a safe pattern storage repository; and
update the pattern matching engine, wherein updating the pattern matching engine further causes the pattern matching engine to determine the one or more additional image pattern statuses based on the image pattern status for the second potentially malicious image pattern.

### Claim 4 (depends on 1)

The computing platform of claim 1, wherein the instructions, when executed, cause the computing platform to monitor interactions between the first user device and the cloud platform by:
receiving interaction information from a custom agent previously installed by the computing platform on a virtual resource of the cloud platform.

### Claim 5 (depends on 1)

The computing platform of claim 1, wherein the real-time interaction information comprises at least one of:
network traffic packet information;
network flow information;
network authentication information; or
processing parameters of the first user device.

### Claim 6 (depends on 1)

The computing platform of claim 1, wherein the instructions, when executed, cause the computing platform to generate the first potentially malicious image pattern by using a consolidation unit, wherein the consolidation unit is configured to consolidate interaction information corresponding to the first user device and interaction information corresponding to the cloud platform into an image pattern comprising at least one of:
a heat map based on a portion of the real-time interaction information;
a graphical representation of the portion of the real-time interaction information; or
a vector representation of the portion of the real-time interaction information.

### Claim 7 (depends on 1)

The computing platform of claim 1, wherein the instructions, when executed, cause the computing platform to initiate the at least one security action by:
generating a security alert comprising a signature of a known botnet corresponding to at least one historical malicious image pattern of the plurality of historical malicious image patterns; and
outputting the security alert, wherein outputting the security alert comprises at least one of:
sending a security report comprising the security alert to a cybersecurity platform; or
causing the first user device to display the security alert.

### Claim 8 (depends on 1)

The computing platform of claim 1, wherein the instructions, when executed, cause the computing platform to initiate the at least one security action by:
segmenting the first user device from the cloud platform;
applying one or more cybersecurity rules to the first user device;
determining, based on applying the one or more cybersecurity rules, whether malicious software associated with the first potentially malicious image pattern has been neutralized; and
reintroducing, based on a determination that the malicious software has been neutralized, the first user device to the cloud platform.

### Claim 9 (depends on 1)

The computing platform of claim 1, wherein the real-time interaction information is further generated based on monitoring interactions between a plurality of additional user devices and the cloud platform.

### Claim 10

A method comprising:
at a computing device comprising at least one processor, a communication interface, and memory:
training, based on a training set of historical image patterns stored in at least one pattern storage repository, a pattern matching engine to generate image pattern statuses, wherein the training set comprises a plurality of historical malicious image patterns and a plurality of historical safe image patterns, wherein a given historical image pattern comprises historical interaction information for a cloud-based network;
automatically generating real-time interaction information based on monitoring interactions between a first user device and a cloud platform;
generating a first potentially malicious image pattern based on the real-time interaction information;
inputting the first potentially malicious image pattern into the pattern matching engine to generate an image pattern status for the first potentially malicious image pattern, wherein generating the image pattern status comprises comparing the first potentially malicious image pattern to at least one historical image pattern;
identifying, based on the comparing, that the first potentially malicious image pattern is a malicious image pattern, and in response:
storing the first potentially malicious image pattern at a malicious pattern storage repository; and
initiating at least one security action responsive to malicious image patterns; and
updating, using the first potentially malicious image pattern and the image pattern status for the first potentially malicious image pattern, the pattern matching engine, wherein updating the pattern matching engine causes the pattern matching engine to determine one or more additional image pattern statuses based on the image pattern status for the first potentially malicious image pattern.

### Claim 11 (depends on 10)

The method of claim 10, further comprising:
retrieving at least one signature of a known botnet from a cybersecurity platform;
inputting the at least one signature into a pattern creation engine to generate at least one known malicious image pattern; and
storing the at least one known malicious image pattern at the at least one pattern storage repository.

### Claim 12 (depends on 10)

The method of claim 10, further comprising:
generating a second potentially malicious image pattern based on the real-time interaction information;
determining an image pattern status for the second potentially malicious image pattern based on inputting the second potentially malicious image pattern into the pattern matching engine;
storing, based on a determination that the second potentially malicious image pattern is a safe image pattern, the second potentially malicious image pattern at a safe pattern storage repository; and
updating the pattern matching engine, wherein updating the pattern matching engine further causes the pattern matching engine to determine the one or more additional image pattern statuses based on the image pattern status for the second potentially malicious image pattern.

### Claim 13 (depends on 10)

The method of claim 10, wherein the generating the first potentially malicious image pattern comprises using a consolidation unit, wherein the consolidation unit is configured to consolidate interaction information corresponding to the first user device and interaction information corresponding to the cloud platform into an image pattern comprising at least one of:
a heat map based on a portion of the real-time interaction information;
a graphical representation of the portion of the real-time interaction information; or
a vector representation of the portion of the real-time interaction information.

### Claim 14 (depends on 10)

The method of claim 10, wherein the initiating the at least one security action comprises:
generating a security alert comprising a signature of a known botnet corresponding to at least one historical malicious image pattern of the plurality of historical malicious image patterns; and
outputting the security alert, wherein outputting the security alert comprises at least one of:
sending a security report comprising the security alert to a cybersecurity platform; or
causing the first user device to display the security alert.

### Claim 15 (depends on 10)

The method of claim 10, wherein the initiating the at least one security action comprises:
segmenting the first user device from the cloud platform;
applying one or more cybersecurity rules to the first user device;
determining, based on applying the one or more cybersecurity rules, whether malicious software associated with the first potentially malicious image pattern has been neutralized; and
reintroducing, based on a determination that the malicious software has been neutralized, the first user device to the cloud platform.

### Claim 16

One or more non-transitory computer-readable media storing instructions that, when executed by a computing platform comprising at least one processor, a communication interface, and memory, cause the computing platform to:
train, based on a training set of historical image patterns stored in at least one pattern storage repository, a pattern matching engine to generate image pattern statuses, wherein the training set comprises a plurality of historical malicious image patterns and a plurality of historical safe image patterns, wherein a given historical image pattern comprises historical interaction information for a cloud-based network;
automatically generate real-time interaction information based on monitoring interactions between a first user device and a cloud platform;
generate a first potentially malicious image pattern based on the real-time interaction information;
input the first potentially malicious image pattern into the pattern matching engine to generate an image pattern status for the first potentially malicious image pattern, wherein generating the image pattern status comprises comparing the first potentially malicious image pattern to at least one historical image pattern;
identify, based on the comparing, that the first potentially malicious image pattern is a malicious image pattern, and in response:
store the first potentially malicious image pattern at a malicious pattern storage repository; and
initiate at least one security action responsive to malicious image patterns; and
update, using the first potentially malicious image pattern and the image pattern status for the first potentially malicious image pattern, the pattern matching engine, wherein updating the pattern matching engine causes the pattern matching engine to determine one or more additional image pattern statuses based on the image pattern status for the first potentially malicious image pattern.

### Claim 17 (depends on 16)

The one or more non-transitory computer-readable media of claim 16 storing instructions that, when executed, further cause the computing platform to:
retrieve at least one signature of a known botnet from a cybersecurity platform;
input the at least one signature into a pattern creation engine to generate at least one known malicious image pattern; and
store the at least one known malicious image pattern at the at least one pattern storage repository.

### Claim 18 (depends on 16)

The one or more non-transitory computer-readable media of claim 16 storing instructions that, when executed, further cause the computing platform to:
generate a second potentially malicious image pattern based on the real-time interaction information;
determine an image pattern status for the second potentially malicious image pattern based on inputting the second potentially malicious image pattern into the pattern matching engine;
store, based on a determination that the second potentially malicious image pattern is a safe image pattern, the second potentially malicious image pattern at a safe pattern storage repository; and
update the pattern matching engine, wherein updating the pattern matching engine further causes the pattern matching engine to determine the one or more additional image pattern statuses based on the image pattern status for the second potentially malicious image pattern.

### Claim 19 (depends on 16)

The one or more non-transitory computer-readable media of claim 16 storing instructions that, when executed, further cause the computing platform to initiate the at least one security action by:
generating a security alert comprising a signature of a known botnet corresponding to at least one historical malicious image pattern of the plurality of historical malicious image patterns; and
outputting the security alert, wherein outputting the security alert comprises at least one of:
sending a security report comprising the security alert to a cybersecurity platform; or
causing the first user device to display the security alert.

### Claim 20 (depends on 16)

The one or more non-transitory computer-readable media of claim 16 storing instructions that, when executed, further cause the computing platform to initiate the at least one security action by:
segmenting the first user device from the cloud platform;
applying one or more cybersecurity rules to the first user device;
determining, based on applying the one or more cybersecurity rules, whether malicious software associated with the first potentially malicious image pattern has been neutralized; and
reintroducing, based on a determination that the malicious software has been neutralized, the first user device to the cloud platform.
