Crow Search Freeman Chain Code (CS-FCC) feature extraction algorithm for handwritten character recognition
With so many algorithms developed to improve classification accuracy, interest in feature extraction in Handwritten Character Recognition (HCR) has increased. In this research, a Crow Search Algorithm (CSA)-based metaheuristic strategy for feature extraction in HCR was developed. The data representa...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English English |
| Published: |
Institute of Electrical and Electronics Engineers Inc.
2023
|
| Subjects: | |
| Online Access: | http://umpir.ump.edu.my/id/eprint/40313/1/Crow%20search%20Freeman%20Chain%20Code%20%28CS-FCC%29%20feature.pdf http://umpir.ump.edu.my/id/eprint/40313/2/Crow%20Search%20Freeman%20Chain%20Code%20%28CS-FCC%29%20feature%20extraction%20algorithm%20for%20handwritten%20character%20recognition_ABS.pdf http://umpir.ump.edu.my/id/eprint/40313/ https://doi.org/10.1109/ICSECS58457.2023.10256286 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | With so many algorithms developed to improve classification accuracy, interest in feature extraction in Handwritten Character Recognition (HCR) has increased. In this research, a Crow Search Algorithm (CSA)-based metaheuristic strategy for feature extraction in HCR was developed. The data representation method employed was Freeman Chain Code (FCC). The fundamental issue with using FCC to represent a character is that the outcomes of the extractions depend on the starting points that changed the chain code's route length. The shortest route length and least amount of computational time for HCR were found using the metaheuristic technique via CSA, which was suggested as a solution to this issue. The suggested CS-FCC extraction algorithm's computation durations and route lengths serve as performance indicators. The algorithm experiments are carried out using the chain code representation created from previous research of the Centre of Excellence for Document Analysis and Recognition (CEDAR) dataset, which consists of 126 upper-case letter characters. According to the results, the proposed CS-FCC has a route length of 1880.28 and only takes 1.10 seconds to solve the entire set of character images. |
|---|