Local AI Content Filtering for Mobile Display Screens

Solution Overview

Problem

Existing content filtering systems for mobile devices rely heavily on external servers, which can be resource-intensive, inefficient, and unable to monitor offline content, leading to increased bandwidth and battery usage, as well as potential encryption issues and cumbersome data transfer.

Innovation Solution

A method and system for local content filtering on mobile devices that samples and preprocesses display content using local resources, employing artificial intelligence and low computational cost methodologies to detect objectionable content, with the option to send ambiguous or objectionable content to an external server for further processing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If content filtering is performed using external servers, then filtering capability is provided, but bandwidth usage and battery consumption increase

Engineering Contradiction:

Improvecontent filtering capabilityVSAvoidbattery consumption

Core Design Contradiction:

ReliabilityVSUse of energy by moving object

Solution Approach 1:

The mobile device performs content filtering operations locally using its own processor and stored software, enabling the system to serve itself without requiring constant external server assistance. This self-service approach reduces network dependency and energy consumption associated with data transmission.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent extracts the essential filtering functionality from the external server environment and implements it locally on the mobile device. By taking out the core processing tasks from the server and placing them on the device, the system reduces bandwidth usage while maintaining filtering capability.

Inventive Principle:

Principle #2Taking out (Extraction)

2Reliability

If content filtering is performed using external servers, then filtering capability is provided, but network bandwidth usage increases

Engineering Contradiction:

Improvecontent filtering capabilityVSAvoidbandwidth usage

Core Design Contradiction:

ReliabilityVSLoss of energy

Solution Approach 1:

The mobile device performs content filtering operations locally using its own processor and stored software, enabling the system to serve itself without requiring constant external server assistance. This self-service approach reduces network dependency and energy consumption associated with data transmission.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent extracts the essential filtering functionality from the external server environment and implements it locally on the mobile device. By taking out the core processing tasks from the server and placing them on the device, the system reduces bandwidth usage while maintaining filtering capability.

Inventive Principle:

Principle #2Taking out (Extraction)

3Adaptability or versatility

If comprehensive content analysis is performed locally, then offline filtering is enabled, but device resources are consumed

Engineering Contradiction:

Improveoffline filtering capabilityVSAvoidlocal processing requirements

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The content filtering process is segmented into multiple stages: sampling/display capture, preprocessing with locally stored software, and AI-based analysis. This segmentation allows the system to distribute processing tasks appropriately, enabling offline capability while managing device resource consumption through progressive analysis stages.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system performs preliminary preprocessing of captured display content using locally stored software before applying more resource-intensive AI analysis. This preliminary action prepares the data in advance, making the subsequent filtering process more efficient and reducing the overall computational burden on the device.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS20230162337A1Real time local filtering of on-screen images

Publication Date: 2023.05.25 NETABPARK

AI summary

The present invention relates to a system and method for detecting inappropriate content on a device and filtering content on a variety of media. Inappropriate content is detected by sampling a display of the device to produce a sample, preprocessing the sample using a local processor and locally stored software to determine if the sample is a likely candidate to include objectionable content, and if found to be a likely candidate, analyzing the sample using an artificial intelligence routine running on said local processor;