Focus Detection Device Spatial Addition Similarity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing focus detection methods face challenges in maintaining ranging precision, especially when dealing with different subject patterns, as they often result in focus detection errors or impossibility due to low contrast values and incorrect spatial addition of pixel data.
Innovation Solution
A focus detection device and method that calculate the degree of similarity between pixel signal rows from different directions and perform spatial addition only when the similarity is high, generating a spatial addition pixel signal for phase difference detection, thereby improving focus detection precision.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If spatial addition is performed on pixel rows with different subject patterns, then the signal level increases, but ranging precision deteriorates
Solution Approach 1:
The patent changes the parameter of pixel row selection by calculating similarity between pixel rows and selectively adding only those with high similarity. This parameter change (from adding all pixel rows to adding only similar ones) resolves the contradiction by maintaining signal level improvement while preventing ranging precision deterioration through similarity-based filtering
Solution Approach 2:
The patent introduces a feedback mechanism where the similarity calculation result determines whether spatial addition should be performed. The system calculates similarity between pixel rows, uses this result as feedback to decide on addition, thereby preventing deterioration of ranging precision while maintaining signal level improvement when appropriate
2Quantity of substance
If time series addition is performed on moving subjects, then the signal level increases, but focus detection errors occur
Solution Approach 1:
The patent changes the parameter of addition timing by calculating similarity between pixel rows and selectively performing spatial addition only when similarity exceeds a threshold. This prevents time series addition errors on moving subjects while maintaining signal level improvement when subjects are stationary or similar
Solution Approach 2:
The patent makes the addition method dynamic by adjusting based on similarity calculation results. The system adapts between spatial addition and time series addition based on whether pixel rows are similar, thereby preventing focus detection errors on moving subjects while maintaining signal level improvement when appropriate
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach effectively prevents deterioration in ranging precision by determining the appropriate method of pixel data addition based on similarity, ensuring accurate focus detection even with varying subject patterns.
Implementation Method 1
an image sensor having a plurality of pixel sections that subject light flux from a subject that has passed through a photographing lens to pupil division, and that are made up of pairs of pixels that respectively receive the light flux that has been subjected to pupil division, and that outputs pairs of pixel signal rows corresponding to pupil division by photoelectric conversion
Data Source
AI summary
A focus detection device, comprising an image sensor having pixel sections that subject light flux that has passed through a photographing lens to pupil division, and that are made up of pairs of pixels that respectively receive the light flux, and processors, wherein the one or more processors calculate degree of similarity between pixel signal rows that are positioned in a direction that is different to the pupil division direction, and are constituted by pixel signals of a plurality of pixels corresponding to at least one of the pairs, and if the degree of similarity is higher that a specified value, generate a spatial addition pixel signal by adding the pairs of pixel signals of the pixel signal row in a direction that is different to the pupil division direction, for each respective pair, and perform phase difference detection based on this spatial addition pixel signal that has been generated.