Autofocus Curve Fitting Reduces Computation Time

Solution Overview

Problem

Current digital camera autofocus systems are inefficient, taking longer to determine the optimal lens position and often using redundant positions, leading to wasted computation time and power resources, resulting in suboptimal image capture.

Innovation Solution

A method and system that determine an initial lens position dataset, calculate scores, and use interpolation and extrapolation with polynomial curve fitting analysis to estimate the optimal focus position, reducing the number of required lens positions and improving focus accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional autofocus methods are used to determine optimal lens position, then focus accuracy is achieved, but computation time and power consumption increase significantly

Engineering Contradiction:

Improvefocus accuracyVSAvoidcomputation time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary action by using the image sensor to capture preliminary focus information during normal image capture, which is then used to determine the optimal lens position without requiring additional time-consuming external sensor measurements or exhaustive lens position testing

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent applies copying by using the image sensor (which captures the final image) to also perform focus measurement functions, effectively copying the focus detection capability from dedicated external sensors to the existing image sensor, thereby eliminating redundant measurement processes

Inventive Principle:

Principle #26Copying

2Measurement precision

If traditional autofocus methods are used to determine optimal lens position, then focus accuracy is achieved, but power consumption increases due to redundant lens positions

Engineering Contradiction:

Improvefocus accuracyVSAvoidpower consumption

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The system performs preliminary action by using the image sensor to capture preliminary focus information during normal image capture, which is then used to determine the optimal lens position without requiring additional time-consuming external sensor measurements or exhaustive lens position testing

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent applies copying by using the image sensor (which captures the final image) to also perform focus measurement functions, effectively copying the focus detection capability from dedicated external sensors to the existing image sensor, thereby eliminating redundant measurement processes

Inventive Principle:

Principle #26Copying

3Measurement precision

If more lens positions are tested to ensure accurate focus, then focus accuracy improves, but computation time and power resources are wasted

Engineering Contradiction:

Improvefocus accuracyVSAvoidimage capture speed

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The system performs preliminary action by using the image sensor to capture preliminary focus information during normal image capture, which is then used to determine the optimal lens position without requiring additional time-consuming external sensor measurements or exhaustive lens position testing

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent applies copying by using the image sensor (which captures the final image) to also perform focus measurement functions, effectively copying the focus detection capability from dedicated external sensors to the existing image sensor, thereby eliminating redundant measurement processes

Inventive Principle:

Principle #26Copying

Data Source

PatentUS9621780B2Method and system of curve fitting for common focus measures

Publication Date: 2017.04.11 NVIDIA CORP

AI summary

An efficient method and system for estimating an optimal focus position for capturing an image are presented. Embodiments of the present invention initially determine an initial lens position dataset. Then, scores are calculated for each value of the initial lens position dataset producing a plurality of scores. Embodiments of the present invention then determine an optimum focus position through interpolation and extrapolation by relating the initial lens position dataset to the score dataset, in which the score dataset comprises of the plurality of scores.