Event Sensor Autofocus Using Light Intensity Variations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autofocus systems suffer from slow response times, low dynamic range, and require user intervention for AF window selection due to limitations in conventional image sensors.
Innovation Solution
The use of event sensing pixels to capture light intensity variations, allowing for faster scene change detection and interest region selection without the need for color information or absolute light intensity detection, enabling a more efficient autofocus method with reduced memory and power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If conventional image sensors are used for autofocus, then color information and absolute light intensity can be detected, but response time is slow and dynamic range is low
Solution Approach 1:
The patent extracts only the necessary information (light intensity variations) from the scene while discarding unnecessary information (color data, absolute intensity values). Event sensing pixels are designed to detect only changes in light intensity, not absolute values or color information, thereby achieving faster response and higher dynamic range while losing only the information not required for autofocus functionality.
2Quantity of substance
If conventional image sensors are used for autofocus, then complete image data is captured, but memory buffer and computation requirements increase
Solution Approach 1:
The system extracts only event data (pixels with light intensity changes) from the scene rather than capturing complete image frames. This selective extraction dramatically reduces the quantity of data that needs to be stored in memory buffers and processed computationally, while still providing sufficient information for accurate autofocus operation.
Solution Approach 2:
Instead of processing complete image frames (excessive action), the system processes only the partial information necessary for autofocus - specifically, event data indicating which pixels have changed in light intensity. This partial processing approach reduces computational requirements and memory buffer needs while maintaining autofocus accuracy.
3Reliability
If conventional image sensors are used for autofocus, then standard imaging is achieved, but power consumption is high
Solution Approach 1:
The event sensing pixels extract only change information from the optical scene, avoiding the need to continuously capture and process complete image frames. This extraction approach maintains reliable autofocus functionality while significantly reducing power consumption compared to conventional image sensors that must continuously read out and process full frame data.
Solution Approach 2:
Instead of continuous frame capture, the event sensor operates by detecting discrete events (light intensity changes) that occur periodically or asynchronously. This event-driven periodic action reduces average power consumption while maintaining imaging capability for autofocus, as the sensor only activates processing when actual changes occur in the scene.
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 results in a faster, more accurate, and user-friendly autofocus system with higher dynamic range and reduced computational requirements, enhancing user experience by eliminating the need for user intervention and improving scene-change detection accuracy.
Implementation Method 1
a plurality of event sensing pixels, configured to capture event data of the targeted scene, wherein the event data indicates which pixels of the event sensing pixels have changes in light intensity
Data Source
AI summary
A focus method and an image sensing apparatus are disclosed. The method includes capturing, by a plurality of event sensing pixels, event data of a targeted scene, wherein the event data indicates which pixels of the event sensing pixels have changes in light intensity, accumulating the event data for a predetermined time interval to obtain accumulated event data, determining whether a scene change occurs in the targeted scene according to the accumulated event data, obtaining one or more interest regions in the targeted scene according to the accumulated event data in response to the scene change, and providing at least one of the one or more interest regions for a focus operation. The image sensing apparatus comprises a plurality of image sensing pixels, a plurality of event sensing pixels, and a controller configured to perform said method.