Predicted Background Image Generation for Object Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for generating predicted background images for object detection fail to maintain uniform light source environments, leading to inaccurate object detection, especially in outdoor settings where light source conditions change.

Innovation Solution

The method generates a small region panorama image group with overlapping regions, allowing for the creation of a predicted background image with a consistent light source environment, using linearization processing to match the target image's light source conditions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If a predicted background image is synthesized from multiple shot images with different light source environments, then the coverage area and adaptability are improved, but the uniformity of light source environment deteriorates

Engineering Contradiction:

Improveadaptability to different shooting conditionsVSAvoiduniformity of light source environment

Core Design Contradiction:

Adaptability or versatilityVSIllumination intensity

Solution Approach 1:

The patent segments the panorama image into multiple small region images, each captured under different light source environments. By processing and selecting appropriate regions from these segmented images, the system can generate a predicted background image that matches the light source environment of the detection target image, thus resolving the contradiction between adaptability and uniformity.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies local quality by allowing different regions of the predicted background image to be sourced from different small region images, each with its own light source characteristics. This enables the predicted background image to have locally appropriate light source environments that match the corresponding regions of the detection target image.

Inventive Principle:

Principle #3Local quality

2Productivity

If background difference is performed using a predicted background image with non-uniform light source environment, then the processing speed is maintained, but the detection accuracy deteriorates due to intensity differences

Engineering Contradiction:

Improveprocessing speedVSAvoiddetection accuracy

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The patent changes the light source environment parameter of the predicted background image by selecting and processing small region images with matching light source characteristics. This parameter matching eliminates intensity differences between the predicted background image and detection target image, thereby improving detection accuracy while maintaining processing efficiency.

Inventive Principle:

Principle #35Parameter changes

3Adaptability or versatility

If light source environment changes after synthesis of predicted background image, then the adaptability to changing conditions is improved, but the consistency between detection target image and predicted background image deteriorates

Engineering Contradiction:

Improveadaptability to light source changesVSAvoidconsistency of light source environment

Core Design Contradiction:

Adaptability or versatilityVSStability of the object's composition

Solution Approach 1:

The patent performs preliminary action by capturing multiple small region images under different light source environments before the actual detection. When the detection target image is acquired, the system selects and processes the appropriate pre-captured small region image that matches the current light source environment, thereby maintaining consistency despite environmental changes.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent introduces dynamics by making the predicted background image adaptable to changing light source environments. The system dynamically selects and processes small region images based on the light source conditions of the detection target image, allowing the predicted background image to change according to environmental conditions while maintaining consistency.

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS7747104B2Image generating method, object detecting method, object detecting equipment, and image generating program

Publication Date: 2010.06.29 PANASONIC INTELLECTUAL PROPERTY CORP OF AMERICA

AI summary

A small region panorama image group (I1P, I2P, I3P) is generated in advance. The small region panorama image group (I1P, I2P, I3P) includes images including a common shooting range, being composed of a plurality of common partially overlapped regions (A, B, C, D), and having independency from each other in each partially overlapped region (A, B, C, D). With the use of the small region panorama image group (I1P, I2P, I3P), a predicted background image is generated in which light source environment is the same as that of a shot target image.