Target Recognition Based on Improved Fourier Descriptors

To address the problems of low recognition accuracy, slow computing speed, large memory consumption, high requirements for hardware facilities of traditional image target recognition algorithms and the deep learning method data set collection difficulties. In this paper, we propose an improved fast...

Full description

Saved in:
Bibliographic Details
Main Authors: Huang, Mengtao, Yang, Tao, Fu, Chen, Fan, Xinfeng
Format: Conference Proceeding
Language:English
Subjects:
Fourier descriptors
Fourier transforms
Image recognition
Libraries
Memory management
Object detection
OpenCV
Shape
Target recognition
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:To address the problems of low recognition accuracy, slow computing speed, large memory consumption, high requirements for hardware facilities of traditional image target recognition algorithms and the deep learning method data set collection difficulties. In this paper, we propose an improved fast Fourier descriptors that can recognize target patterns quickly and accurately. The method first performs image enhancement and other pre-processing on the acquired image. Then uses the butterfly operation algorithm to accelerate the Fourier transform, calculates the Fourier descriptors of the target shape and improves the operation speed by reducing the number of key points of the target shape. Finally, the Fourier descriptors contours of the pattern to be detected and the template pattern are matched to determine the authenticity of the target recognition, and the OpenCV library is used for experimental simulation. The experimental results show that the method can accurately identify the target pattern and improve the recognition speed.
ISSN:2688-0938
DOI:10.1109/CAC57257.2022.10055443