Movement Path Inference Using Shape and Position Similarity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image analysis techniques struggle to correctly recognize movement types when there are two or more learned movement path curves that are similar in shape, due to the inability to differentiate between them accurately.
Innovation Solution
An inference apparatus that computes shape and position similarities between learned and observed movement path curves, utilizing speed and distribution coefficients to enhance recognition accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If only shape similarity comparison is used to recognize movement types, then the system is robust against position differences, but it cannot correctly distinguish between multiple learned curves with similar shapes
Solution Approach 1:
The patent segments the similarity assessment into two distinct components: shape similarity comparison (which is robust to position differences) and position similarity comparison (which provides discrimination power). By dividing the recognition task into these separate evaluation dimensions, the system can leverage both the robustness of shape-based comparison and the discriminative power of position-based comparison to achieve accurate movement type recognition even when curves have similar shapes.
2Measurement precision
If position similarity comparison is added to shape similarity comparison, then movement type recognition accuracy improves, but the computational complexity increases
Solution Approach 1:
The patent implements a two-stage comparison process where shape similarity is evaluated first as a primary filter, and position similarity is computed only when needed (e.g., when shape similarity alone is insufficient for discrimination). This partial application of the more computationally intensive position similarity calculation reduces overall computational complexity while still achieving accurate recognition when required.
Data Source
AI summary
A shape similarity computation unit (2021) computes, as a shape similarity, a similarity in shape between a learning movement path curve which is a movement path curve obtained through learning and an observation movement path curve which is a movement path curve obtained through observation. A position similarity computation unit (2022) computes, as a position similarity, a similarity in position between the learning movement path curve and the observation movement path curve when the learning movement path curve and the observation movement path curve are placed in a same coordinate space. A conformity computation unit (203) computes a conformity between the learning movement path curve and the observation movement path curve, using the shape similarity and the position similarity.