Moving Object Detection in Focus Stacking via Contrast and Luminance Segmentation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional focus stacking techniques deteriorate when a moving object is present in the images, as they incorrectly identify non-moving objects with color or luminance differences as moving objects.

Innovation Solution

An apparatus that acquires evaluation values related to changes in color and luminance from multiple images of different focus positions and uses these values to detect moving objects, preventing incorrect identification by analyzing contrast changes and generating a color/luminance difference map and contrast change map to exclude moving objects from the composite image.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If areas with great differences in color or luminance are extracted to identify moving objects, then moving objects can be excluded from focus stacking, but non-moving objects with color or luminance differences may be wrongly recognized as moving objects

Engineering Contradiction:

Improveaccuracy of moving object detectionVSAvoidprecision of moving object area identification

Core Design Contradiction:

ReliabilityVSMeasurement precision

Solution Approach 1:

The patent divides the detection process into two separate segmentation stages: first extracting areas with color/luminance differences, then extracting areas with contrast differences. By segmenting the detection into multiple criteria, the system can cross-validate results and reduce false positives where non-moving objects are mistakenly identified as moving objects.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces contrast difference as an intermediary criterion between color/luminance difference detection and final moving object identification. This intermediary step acts as a filter that validates whether detected areas are truly moving objects by checking if they also exhibit contrast differences characteristic of motion, thereby reducing false identification.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Reliability

If multiple evaluation criteria are used to improve moving object detection accuracy, then false identification is reduced, but detection complexity increases

Engineering Contradiction:

Improveaccuracy of moving object detectionVSAvoidcomplexity of detection process

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The detection process is segmented into distinct modular steps: color/luminance difference extraction, contrast difference extraction, and composite image generation. Each module performs a specific function independently, making the overall complex process manageable and allowing for optimized implementation of each segment without increasing overall system complexity.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS12106490B2Apparatus for detecting moving object, method for detecting moving object, and non-transitory computer-readable storage medium

Publication Date: 2024.10.01 CANON KK

AI summary

An apparatus includes at least one memory configured to store instructions, and at least one processor in communication with the at least one memory and configured to execute the instructions to acquire a first evaluation value related to a change in color or luminance from a plurality of images different in focus position, acquire a second evaluation value related to a change in contrast from the plurality of images, and detect a moving object from the plurality of images based on the first evaluation value and the second evaluation value.