Image Pixel Determination Using Brightness Thresholds and Edge Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image determining methods rely solely on brightness thresholds, which are inadequate for accurately identifying specific image pixels and computing object coordinates, as they fail to consider other parameters and are sensitive to object size and type variations.
Innovation Solution
An image determining method that utilizes brightness values and additional parameters to identify specific image pixels by defining initial and subsequent pixel ranges, allowing for more accurate object coordinate computation, including the use of a reading circuit to scan images and compute the gravity center.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a high brightness threshold value is used to determine specific image pixels, then false positives are reduced, but edge pixels with lower brightness values are excluded from detection
Solution Approach 1:
The patent segments the image processing into multiple stages: first identifying candidate pixels using brightness threshold, then expanding to adjacent pixels within a predetermined range, and finally determining the complete object boundary through iterative processing. This segmentation allows different criteria to be applied at different stages, resolving the contradiction between high threshold accuracy and complete boundary detection.
Solution Approach 2:
The patent performs preliminary identification of high-brightness pixels as seed points before completing the full object boundary detection. These preliminary identified pixels serve as starting points for subsequent expansion operations, enabling the system to first reliably locate object centers and then systematically capture the complete boundary including low-brightness edge regions.
2Measurement precision
If a low brightness threshold value is used to determine specific image pixels, then more edge pixels are included, but normal pixels are incorrectly identified as specific image pixels
Solution Approach 1:
The patent applies different brightness threshold criteria to different spatial regions. High-brightness thresholds are applied at seed point locations, while lower thresholds are applied during expansion to adjacent pixels within the predetermined range. This local differentiation allows the system to maintain high reliability at object centers while achieving complete boundary detection at edges.
Solution Approach 2:
The patent dynamically adjusts the effective threshold during the detection process. The threshold is not fixed but varies depending on the processing stage and spatial position, being stricter for initial pixel identification and more permissive for boundary expansion, thereby adapting to the different requirements of reliability and completeness at different phases.
3Productivity
If only brightness values are used for pixel determination, then the method is simple and fast, but it cannot accommodate variations in object size and type
Solution Approach 1:
The patent creates a universal detection framework that can handle various object types and sizes through a standardized multi-stage process. The predetermined pixel range expansion and iterative boundary following mechanisms work consistently across different objects, making the system adaptable to diverse targets while maintaining processing efficiency through algorithmic standardization.
Solution Approach 2:
The patent introduces additional parameters beyond brightness values, including spatial relationships (adjacent pixels within predetermined ranges), positional information (row and column coordinates), and iterative expansion parameters. These parameter changes enable the system to adapt to different object characteristics while maintaining a unified processing approach that preserves computational efficiency.
4Measurement precision
If the brightness threshold is adjusted to capture all edge pixels, then boundary completeness improves, but the determination accuracy decreases due to inclusion of normal pixels
Solution Approach 1:
The patent introduces an intermediary expansion process between threshold-based pixel identification and final boundary determination. The predetermined pixel range expansion acts as an intermediary step that bridges high-brightness seed pixels and low-brightness edge pixels, allowing the system to achieve complete boundary detection without directly lowering the initial threshold, thus maintaining determination accuracy while improving boundary completeness.
Data Source
AI summary
An image determining method for scanning an image and determining specific image pixels of a specific image. The method comprises: determining at least one pixel in a first row having brightness value larger than a threshold value as the specific image pixel; defining a leftmost pixel and a rightmost pixel of the specific image pixel in the first row as a leftmost edge and a rightmost edge of a first specific image range; and defining a second specific image range in a second row of the image, which is next to the first row. Column positions of a leftmost edge and a rightmost edge of the second specific image range are respectively the same with column positions of the leftmost edge and the rightmost edge of the first specific image range. Via this method, the determining for the specific image pixels is more accurate.