Fast Iris localization using Haar-like features and AdaBoost algorithm
- PDF / 2,608,786 Bytes
- 24 Pages / 439.37 x 666.142 pts Page_size
- 10 Downloads / 213 Views
Fast Iris localization using Haar-like features and AdaBoost algorithm Yi-Nan Lin 1 & Tsang-Yen Hsieh 1 & Jr-Jen Huang 1 & Cheng-Ying Yang 2 & Victor R. L. Shen 3,4 & Hai Hoang Bui 1 Received: 16 May 2019 / Revised: 25 November 2019 / Accepted: 30 March 2020 # Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract
Traditional iris recognition methods, which are still preferred against artificial intelligence (AI) approaches in practical applications, are often required to capture high-grade iris samples by an iris scanner for accurate subsequent processing. To reduce the system cost for mass deployment of iris recognition, pricey scan devices can be replaced by the average quality cameras combined with additional processing algorithm. In this paper, we propose a Haar-like-feature-based iris localization method to quickly detect the location of human iris in the images captured by low-cost cameras for the ease of post-processing stages. The AdaBoost algorithm was chosen as a learning method for training a cascade classifier using Haar-like features, which was then utilized to detect the iris position. The experimental results have shown acceptable accuracy and processing speed for this novel cascade classifier. This achievement stimulates us to implement this novel capturing device in our iris recognition. Keywords Object detection . Iris localization . Haar-like features . AdaBoost algorithm . Cascade classifier
1 Introduction Despite the domination of artificial intelligence (AI) - based methods in modern scientific researches, the traditional approaches still have a certain degree of impact on practical object recognition. Complex recognition tasks like iris recognition require a large-scale neural network and a large number of training datasets to achieve decent accuracy [17], which seems to be inappropriate for general iris recognition with limited resources. However, the traditional recognition methods also suffer from several negative impact factors such as low-resolution iris samples, off-angle or small-size iris in the samples, and the occlusion of eyelids, eyelashes,
* Victor R. L. Shen [email protected]; [email protected] Extended author information available on the last page of the article
Multimedia Tools and Applications
glasses, wrinkles, and specular. These unfavorable impacts often cause incorrect iris boundaries detection in the segmentation step, leading to the failure of the subsequent steps in the traditional approaches [6]. To solve the poor quality sample problem for low-cost devices, the iris localization algorithm can be applied between the image capturing stage and the iris segmentation stage, which does not increase the computational complexity significantly. The result of this algorithm is a Region of Interest (ROI) which is large enough to fully contain the iris area in the image. Thus, this method is actually similar to common object detection, like face or eye detection. In a particular study, we discovered that performing detection algorithm
Data Loading...
