Design of Chinese Character Coded Targets for Feature Point Recognition Under Motion-Blur Effect

Yun Shi,Liyan Zhang

doi:10.1109/access.2020.3006020

Abstract

Coded targets have been widely used as a type of active visual feature points in fields such as close-range photogrammetry, robot navigation, 3D reconstruction, and augmented reality. However, coded targets in degraded images, such as images with motion blur effect, are hard to recognize. To this end, a set of novel Chinese character coded targets (CCTs) is designed and tested. A CCT is a square visual marker, shown as a relative small circular feature overlaid in the middle of a square Chinese character. A white circular ring is embedded in a black circle concentrically to serve as the circular feature, which facilitates extraction of the center point of the marker for localization. Whereas the distinctive peripheral Chinese character appearance of each CCT is utilized for identification. By synthesizing simulated CCTs with different degrees of motion blur and various postures with the real background images, a Faster Region-based Convolutional Neural Network (Faster R-CNN) is trained to locate and recognize the CCTs in motion blurred images. Experimental results on both artificial and actual motion blurred images demonstrate the superiorities of the designed CCTs as well as the proposed localization and recognition pipeline.

Highlights

Coded targets are widely applied in the fields of machine vision, photogrammetry, augmented reality, and 3D reconstruction, as their unique identity information allows automatic identification and robust image matching
Coded targets can be classified into ring-shaped coded targets [1]–[5], point coded targets [6]–[8], and specially designed coded targets [9]–[14]
A ring-shaped coded target has a circular structure, mainly composed of a circular target at the center and coded bits surrounding it in a concentric ring

Summary

Introduction

Coded targets are widely applied in the fields of machine vision, photogrammetry, augmented reality, and 3D reconstruction, as their unique identity information allows automatic identification and robust image matching. Coded targets can be classified into ring-shaped coded targets [1]–[5], point coded targets [6]–[8], and specially designed coded targets [9]–[14]. A ring-shaped coded target has a circular structure, mainly composed of a circular target at the center and coded bits surrounding it in a concentric ring. An increase in the number of coded bits increases the number of codes available. The ring-shaped coded targets are relative easy to be recognized, but the coding space is usually limited. A point coded target is a digital code based on different distributions of points on a plane. Compared with a ringshaped coded target, the point coded target has a larger coding space but the decoding algorithm is relative complicated

Methods

Findings

Discussion

Conclusion