Genetic Programming Program Generation Accelerating Convergence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing techniques for automatically generating image processing programs using genetic programming are time-consuming, especially when a large number of selectable partial programs are used, leading to prolonged program generation processing times without ensuring high-quality image processing.
Innovation Solution
A program generation apparatus and method that employs genetic programming to optimize the combination of partial programs by tracking fitness level changes and adjusting selection probabilities based on the contribution of each partial program, facilitating faster convergence to high-quality image processing programs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If more selectable partial programs are prepared to improve image processing quality, then the quality of image processing is improved, but the time needed for program generation processing increases
Solution Approach 1:
The patent performs preliminary action by calculating and storing fitness level change amounts for each partial program before the actual genetic programming execution. This pre-computation allows the system to quickly reference which partial programs contribute most to fitness improvement during the evolution process, avoiding time-consuming real-time calculations and enabling faster convergence to high-quality solutions
Solution Approach 2:
The patent implements feedback by using the calculated fitness level change amounts to dynamically adjust selection probabilities of partial programs. The system continuously monitors which partial programs contribute most to fitness improvement and increases their selection probability accordingly, creating a feedback loop that accelerates convergence to high-quality image processing programs without requiring exhaustive exploration of all possible program combinations
2Reliability
If more selectable partial programs are used to achieve high-quality image processing, then the fitness level can reach higher values, but more time is needed for the fitness level to reach a high value
Solution Approach 1:
The system performs preliminary calculation of fitness level change amounts for each partial program before the genetic programming evolution begins. This pre-computation creates a knowledge base that guides the evolution process, allowing the system to make informed selection decisions from the first generation and significantly reducing the number of generations needed to reach high fitness levels
Solution Approach 2:
The patent changes the selection probability parameter dynamically based on fitness level change amounts. By adjusting this parameter according to the pre-calculated contribution of each partial program, the system optimizes the evolution trajectory and achieves high fitness levels faster than traditional genetic programming with uniform selection probabilities
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Time needed to generate an image processing program is shortened. Each time evolution processing is performed, an information recording unit (1b) records increase-decrease information (2a) indicating a type of partial program changed in number between pre- and post-evolution individuals and difference between fitness levels calculated from the pre- and post-evolution individuals in a storage unit (1d). A probability setting unit (1c) determines, per partial program, a contribution level indicating how much a partial program contributes to increasing a fitness level, based on the information (2a) stored in the storage unit (1d) within a predetermined period and sets a selection probability per partial program in such a manner that a partial program having a higher contribution level is given a higher selection probability. When the evolution processing is performed, a generation processing unit (1a) selects a partial program to be included in the post-evolution individual in mutation processing according to the set selection probabilities.