Feature Candidate Pixel Detection Using Coarse Preprocessing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Software implementations of feature detection algorithms in image processing are inefficient due to a large number of conditional branch instructions, leading to high processing time and resource consumption, limiting data level parallelism and execution efficiency on microprocessors.
Innovation Solution
Implementing a preprocessing step, known as coarse detection, to identify feature candidate pixels by comparing a subset of surrounding pixels to a comparison value based on attribute values and threshold values, reducing the number of pixels processed by the feature detection algorithm.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a software implementation of a feature detection algorithm processes all pixels of an image, then comprehensive feature detection is achieved, but processing time and processor resources are significantly consumed
Solution Approach 1:
The patent divides the image processing into two distinct stages: coarse detection that processes the entire image to identify candidate pixels, and fine detection that processes only the candidate pixels. This segmentation reduces the total number of pixels requiring intensive processing while maintaining detection accuracy through the two-stage approach.
Solution Approach 2:
The patent performs preliminary coarse detection on all pixels before conducting fine detection. This preliminary action identifies and flags only those pixels that are likely to be features, preparing a reduced set of candidate pixels for the subsequent fine detection stage, thereby reducing overall processing time.
2Measurement precision
If a software implementation of a feature detection algorithm uses many conditional branch instructions, then detailed pixel analysis is performed, but data level parallelism is limited and execution efficiency decreases
Solution Approach 1:
The patent segments the feature detection algorithm into coarse detection and fine detection stages, each with different computational requirements. The coarse detection stage uses simpler operations that are more amenable to parallel processing, while the fine detection stage applies more complex analysis only to the reduced set of candidate pixels.
Solution Approach 2:
The patent extracts and processes only the relevant subset of pixels (candidate pixels) identified by coarse detection, removing unnecessary processing of non-candidate pixels. This extraction approach reduces the number of conditional branch instructions executed and improves data level parallelism by focusing computational resources on relevant data.
Data Source
AI summary
A system and method of processing an image is disclosed. A particular method of determining whether a particular pixel of an image is a feature candidate includes receiving data corresponding to a subset of a plurality of pixels surrounding the particular pixel. Each of the plurality of pixels may be from the image. The method further includes excluding the particular pixel from consideration as a feature candidate based on a comparison of values of the data to a comparison value. The comparison value may be based on an attribute value of the particular pixel and a threshold attribute value.