Image Structure Detection Using Block Feature Combination

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing image processing methods, such as those using VanderBrug filters, require significant computational resources and time due to the need to apply multiple filters to each pixel, making the detection of structure information inefficient.

Innovation Solution

An information processing apparatus that sets patterns of blocks with multiple pixels, calculates feature amounts of pixel values within these blocks, and detects structure information by combining the feature amounts from these blocks, reducing the computational load and processing time.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If multiple filters are applied to each pixel to detect structure information, then detection precision is improved, but processing time and calculation amount increase

Engineering Contradiction:

Improvestructure information detection precisionVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent divides the image processing into block-level operations instead of pixel-level operations. By setting multiple blocks with different patterns and calculating feature amounts for each block, the system processes groups of pixels simultaneously, reducing the number of individual filter applications needed while maintaining detection precision.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent combines multiple block feature amounts to detect structure information. By merging the results from multiple blocks with different patterns, the system achieves comprehensive structure detection without applying multiple filters to each individual pixel, thus reducing processing time while maintaining precision.

Inventive Principle:

Principle #5Merging (Combining)

2Measurement precision

If multiple filters are applied to each pixel to detect structure information, then detection precision is improved, but calculation amount increases

Engineering Contradiction:

Improvestructure information detection precisionVSAvoidcalculation amount

Core Design Contradiction:

Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent segments the calculation process into block-based feature amount computations. By calculating feature amounts for blocks rather than individual pixels, the system reduces the total number of calculations required while preserving the ability to detect structural features through patterned block analysis.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent creates blocks with different patterns that can detect multiple types of structures simultaneously. Each block pattern serves multiple detection purposes, allowing the system to achieve comprehensive structure information detection with fewer calculations compared to applying multiple specialized filters to each pixel.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS10210422B2Information processing apparatus, information processing method, and program

Publication Date: 2019.02.19 CANON KK

AI summary

In order to shorten a time required to detect structure information of an image, in an image obtained by photographing a subject, feature amounts of pixel values in a block including a plurality of pixels are calculated, and the structure information is detected every (N×N) area of an input image on the basis of a combination of the feature amounts of the pixel values in a plurality of blocks.