Real-time Glare Detection in Dynamic Image Regions

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

The process of capturing images for verification, such as government-issued IDs, is often frustrated by backend rejections due to low image quality, particularly when additional irrelevant image data like supporting surfaces is included, leading to inefficient resource usage in both user devices and backend platforms.

Innovation Solution

A user device capable of real-time image analysis masks irrelevant image data, tracks objects, and determines glare levels to provide feedback and automatically capture high-quality images, optimizing glare detection and resource conservation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If the user device captures and processes the entire preview image including supporting surfaces, then complete scene information is obtained, but computing resources, power consumption, and processing time increase significantly

Engineering Contradiction:

Improvescene informationVSAvoidcomputing resources

Core Design Contradiction:

Loss of informationVSUse of energy by moving object

Solution Approach 1:

The patent segments the image processing task by identifying and isolating the document object from the rest of the scene. The processor determines a document region of interest and processes only that specific area, dividing the full image into processed (document) and unprocessed (background) portions, thereby reducing computational load while preserving essential information

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent extracts the document region from the full preview image by determining boundaries and creating a focused region of interest. This extraction allows the system to process only the relevant document area, removing unnecessary background processing and reducing energy consumption while maintaining document verification quality

Inventive Principle:

Principle #2Taking out (Extraction)

2Reliability

If the user device processes the entire preview image to ensure complete information, then all image data is analyzed, but processing time and computational load increase

Engineering Contradiction:

Improveimage qualityVSAvoidprocessing time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent divides the image into a document region of interest and a background region, processing only the document area for verification purposes. This segmentation maintains reliability by ensuring complete document analysis while reducing processing time by excluding irrelevant background areas from intensive processing

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies partial processing by analyzing only the necessary document region rather than the entire image. This partial action approach maintains sufficient reliability for document verification while significantly reducing processing time and computational resources required

Inventive Principle:

Principle #16Partial or excessive action

3Loss of information

If the user device includes supporting surfaces in the captured image, then context information is preserved, but image quality for verification decreases and resource usage increases

Engineering Contradiction:

Improvecontext informationVSAvoidimage quality

Core Design Contradiction:

Loss of informationVSManufacturing precision

Solution Approach 1:

The patent segments the image to separate the document from supporting surfaces, processing only the document region for verification. This maintains manufacturing precision by ensuring high-quality document analysis while the segmented approach allows context information to exist in the unprocessed background portion without degrading document quality

Inventive Principle:

Principle #1Segmentation

4Measurement precision

If the user device processes all preview images completely, then comprehensive analysis is achieved, but computing and network resources are wasted

Engineering Contradiction:

Improveanalysis accuracyVSAvoidnetwork resources

Core Design Contradiction:

Measurement precisionVSLoss of energy

Solution Approach 1:

The patent extracts and processes only the document region from preview images, achieving comprehensive analysis of the relevant content while excluding unnecessary background data. This extraction approach maintains measurement precision for document verification while reducing network resource consumption by minimizing the amount of data that needs to be transmitted and processed

Inventive Principle:

Principle #2Taking out (Extraction)

Data Source

PatentUS20250005922A1Real-time glare detection inside a dynamic region of an image

Publication Date: 2025.01.02 CAPITAL ONE SERVICES LLC

AI summary

A user device may capture a plurality of preview images that each include image data. The user device may process, in substantially real-time, a preview image to identify an object in the preview image, including determining an outline of the object. The user device may create a mask that hides a portion of the image data associated with an area outside of the outline of the object in one or more of the preview images. Based on determining that a parameter associated with glare of the object in the preview images does not satisfy a threshold, the user device may provide, in substantially real-time, feedback to a user, including an instruction to the user to perform an action with respect to the user device or to the object, and, based on determining that the parameter satisfies the threshold, the user device may automatically capture an image of the object.