Computational Scaling of Shape Similarity That has Potential for Neuromorphic Implementation
Current methods for encoding and comparing shapes are computationally demanding and are not suitable for image processing in small portable devices. Here, we describe a simple scan encoding method for transcribing shape information into a 1-D summary. Summaries were derived from an inventory of unkn...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2018-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/8405528/ |
| Summary: | Current methods for encoding and comparing shapes are computationally demanding and are not suitable for image processing in small portable devices. Here, we describe a simple scan encoding method for transcribing shape information into a 1-D summary. Summaries were derived from an inventory of unknown shapes, and these values were used to scale the degree of similarity of pair combinations. The scale values provided a significant level of prediction of human judgments in a match recognition task, suggesting substantial correspondence with human perception of shape similarity. Similarity scores derived with the Procrustes method did not predict human judgments. |
|---|---|
| ISSN: | 2169-3536 |