Chain Code State Tracing for Round Object Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting round objects in images are inefficient, prone to false positives, and susceptible to occlusion or noise, often requiring excessive computational resources or time.
Innovation Solution
The method involves tracing the contours of chain codes while maintaining specific states, allowing only certain next codes, and using a gradient analysis to determine if an object is round, with extended chain codes and a search for missing contour segments to handle occlusions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the complete contour is traversed to compute the full histogram, then shape identification accuracy is improved, but computation time is wasted on non-round objects
Solution Approach 1:
The patent applies preliminary action by performing early termination of contour traversal when certain conditions are met. Specifically, when the chain code histogram shows uniform distribution or when the aspect ratio indicates circularity, the algorithm stops traversing the contour early, avoiding unnecessary computation on non-round objects while maintaining identification accuracy.
Solution Approach 2:
The patent uses partial action by analyzing only portions of the contour histogram rather than requiring complete traversal. The algorithm computes histogram statistics from partial contour data and uses these to make shape identification decisions, reducing computation time while maintaining sufficient accuracy for distinguishing round from non-round objects.
2Productivity
If chain code histogram analysis is used for shape identification, then simplicity and efficiency are improved, but false-positive results increase
Solution Approach 1:
The patent implements feedback mechanisms that continuously monitor chain code traversal and histogram computation. When the algorithm detects patterns consistent with non-round objects (such as asymmetric histograms or specific aspect ratios), it provides feedback to adjust the identification decision, reducing false positives while maintaining efficiency.
Solution Approach 2:
The patent changes parameters in the chain code analysis by computing additional statistics beyond simple histogram counts, such as aspect ratio calculations and gradient analyses. These parameter changes enable more accurate distinction between round and non-round objects, reducing false positives without significantly impacting processing speed.
3Device complexity
If traditional chain code methods are used, then computational resources are reduced, but performance degrades under occlusion or noise
Solution Approach 1:
The patent applies preliminary action by preparing the chain code data structure and initialization before processing begins. This includes pre-defining the chain code symbols, establishing the histogram arrays, and setting up the traversal state machine, which enables efficient handling of occluded or noisy contours without requiring excessive computational resources during actual processing.
Solution Approach 2:
The patent uses copying by creating a simplified representation of the contour through chain code symbols rather than processing the full pixel data. This copying approach maintains the essential shape information while reducing computational complexity, allowing the system to handle occlusion and noise effectively without requiring substantial computational resources.
Data Source
AI summary
A method for identifying objects in a digital image includes tracing chain codes associated with the contour of the object; a series of states is maintained, and the next chain codes in the contour are accepted only if they comply with allowed chain codes for each state. Certain chain codes trigger a transition into a next state. If a disallowed chain code is encountered, the process halts.