Pattern Detection Result Merging for Real-Time Image Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing pattern discrimination methods face challenges in merging results in real-time applications, leading to potential errors and reduced accuracy due to the need for complete overlap detection, which can be time-consuming and may require stopping merge processing partway through.
Innovation Solution
An information processing apparatus and method that acquires and selects pattern discrimination results, determines their similarity, and merges them based on overlap, allowing for partial merge processing to generate stable results even if stopped prematurely, using a CPU to control the selection, determination, and merging units.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If complete overlap detection is performed to merge pattern discrimination results, then merging accuracy is improved, but processing time increases and real-time performance deteriorates
Solution Approach 1:
The patent applies partial action by performing merge processing only on a predetermined number of pattern discrimination results (e.g., top 3) rather than all detected results. This selective approach maintains sufficient merging accuracy for real-time applications while significantly reducing processing time compared to complete overlap detection of all results.
Solution Approach 2:
The patent segments the pattern discrimination results into two groups: those selected for merging (predetermined number) and those not selected. This segmentation allows the system to focus computational resources on merging the most relevant results while ignoring others, thereby reducing overall processing time while maintaining acceptable accuracy.
2Productivity
If merge processing is stopped partway through to meet real-time requirements, then processing speed is improved, but result stability deteriorates
Solution Approach 1:
The patent performs preliminary selection of pattern discrimination results based on similarity criteria before merging. By pre-selecting the most similar results (e.g., top 3 most similar regions) and preparing them for merging in advance, the system ensures that even if merging is stopped partway through, the results remain stable and reliable because only the most relevant candidates were considered.
Solution Approach 2:
The patent changes the parameter of result selection from considering all detected patterns to considering only the top N most similar patterns. This parameter change (limiting to predetermined number) allows the system to achieve both real-time processing speeds and stable results by focusing computational effort on the most significant matches.
3Measurement precision
If all pattern discrimination results are processed for merging, then comprehensive accuracy is improved, but device complexity and computational load increase
Solution Approach 1:
The patent extracts only the most relevant pattern discrimination results (those with highest similarity) for merging, separating them from the rest. This extraction approach reduces computational load by eliminating unnecessary processing of less relevant results while maintaining comprehensive accuracy for the important cases.
Solution Approach 2:
The patent applies partial action by processing only a subset (predetermined number) of pattern discrimination results for merging rather than all results. This reduces computational load and device complexity while maintaining sufficient accuracy for real-time applications by focusing on the most significant matches.
Data Source
AI summary
There is provided with an information processing apparatus. An acquisition unit acquires a plurality of pattern discrimination results each indicating a location of a pattern that is present in an image. A selection unit selects a predetermined number of pattern discrimination results from the plurality of pattern discrimination results. A determination unit determines whether or not the selected predetermined number of pattern discrimination results are to be merged, based on a similarity of the locations indicated by the predetermined number of pattern discrimination results. A merging unit merges the predetermined number of pattern discrimination results for which it was determined by the determination unit that merging is to be performed. A control unit controls the selection unit, the determination unit, and the merging unit to repeatedly perform respective processes.