Image Sensor Feature Extraction With On-Chip Convolution Pooling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Feature extraction processing in image analysis is time-consuming due to the need for data conversion and transfer, limiting the acceleration of this process.
Innovation Solution
A feature extracting element with a light-receiving substrate and laminated substrates containing convolution and pooling processing units, along with a controlling unit that repeats convolution operations using predetermined filter coefficients, allowing direct processing of pixel values and reducing processing time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If feature extraction is performed by processing captured image data, then feature extraction can be performed, but processing time increases significantly due to image generation and transfer steps
Solution Approach 1:
The invention extracts only the necessary pixel values directly from the light-receiving elements without performing full image generation and transfer. The feature extraction unit reads pixel values directly from the image sensor, eliminating unnecessary image processing steps and reducing processing time while maintaining feature extraction capability.
Solution Approach 2:
The invention divides the image data into blocks and performs feature extraction on segmented regions. The feature extraction unit processes pixel values in a divided manner, allowing parallel processing and reducing overall processing time compared to handling complete images.
2Measurement precision
If image data is processed and transferred for feature extraction, then comprehensive feature analysis can be performed, but device complexity and processing overhead increase
Solution Approach 1:
The invention performs preliminary selection of pixel values at the image sensor level before data leaves the sensor. The feature extraction unit identifies and extracts only the necessary pixel values required for feature analysis, reducing subsequent processing complexity while maintaining accuracy.
Solution Approach 2:
The invention introduces a feature extraction unit as an intermediary between the image sensor and processing systems. This unit acts as a mediator that directly reads pixel values from the sensor and provides extracted features to processing systems, eliminating the need for complete image transfer and reducing system complexity.
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 significantly reduces processing time and eliminates the need for data storage and transfer, enabling faster feature extraction and miniaturization of devices while maintaining processing speed with increasing pixel numbers.
Implementation Method 1
a light-receiving substrate where a plurality of light-receiving elements for photoelectrically converting received light are two-dimensionally arrayed
Data Source
AI summary
An element includes a plurality of light-receiving elements to photoelectrically convert light received from an object, a convolution processing unit to perform convolution operation on signals that are output from the plurality of light-receiving elements, and a pooling processing unit to sample a signal that is output from the convolution processing unit, based on a predetermined condition. The convolution operation of the convolution processing unit and the sampling of the pooling processing unit are repeated.