Image Sensor Edge Extraction via Block Amplitude Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image sensors face challenges in effectively extracting image edges from electrical signals, particularly in recognizing patterns and subjects, due to noise and the need for enhanced edge detection capabilities.
Innovation Solution
An image sensor and method that utilize a pixel array partitioned into blocks, with a converter to convert image signals into digital signals and an image signal processor to add amplitudes, determine edge blocks, and connect direction lines to extract image edges based on signal amplitudes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional image sensors are used for pattern recognition, then basic image capture is achieved, but edge detection accuracy is insufficient due to noise
Solution Approach 1:
The pixel array is divided into multiple blocks, and edge detection is performed block-by-block by comparing adjacent blocks. This segmentation approach allows localized edge detection with reduced noise impact, as each block's edge characteristics are determined independently based on amplitude comparisons with neighboring blocks.
Solution Approach 2:
The patent applies different processing approaches to different regions by comparing signal amplitudes between adjacent blocks. Edge blocks are identified locally where amplitude differences exceed thresholds, and direction lines are determined based on local amplitude relationships, enabling adaptive noise filtering that preserves true edges while suppressing noise.
2Measurement precision
If complex edge detection algorithms are applied to improve edge extraction accuracy, then measurement precision improves, but processing time and power consumption increase
Solution Approach 1:
The patent performs edge detection only on blocks identified as edge blocks based on amplitude comparison thresholds, rather than processing the entire image uniformly. This partial action approach reduces the number of computations required while maintaining edge detection accuracy, as full processing is applied only where edges are likely to occur.
Solution Approach 2:
The patent uses amplitude thresholds as parameters to identify edge blocks and determine direction lines. By changing the threshold parameters adaptively, the system can adjust its sensitivity to edges, improving detection accuracy while controlling processing complexity and time based on the specific image characteristics.
3Use of energy by moving object
If traditional image sensors are used, then basic functionality is maintained, but power consumption is high due to inefficient signal processing
Solution Approach 1:
By segmenting the image into blocks and processing only edge blocks identified through amplitude comparison, the patent reduces the overall computational load. This segmentation enables the system to maintain edge detection functionality while consuming less power, as fewer blocks require full processing.
Solution Approach 2:
The image sensor performs edge detection and direction line determination using its own signal processing capabilities without requiring external processing units. The amplitude comparison method enables the sensor to self-identify edge blocks and extract edges efficiently, improving productivity while reducing power consumption by eliminating the need for additional high-power processing components.
Data Source
AI summary
Provided is an image sensor. The image sensor includes a pixel array including pixels arranged along a first direction and a second direction, and partitioned into blocks, a converter configured to convert image signals outputted from the pixels into digital signals based on an image, and an image signal processor configured to add amplitudes of the digital signals belonging to each of the blocks to determine edge blocks among the blocks, compare the amplitudes of the digital signals to determine directions in which direction lines of the edge blocks are directed, and connect the direction lines to extract an edge of the image.