1
Electrical & Computer Engineering, Iranian Research Organization for Science and Tech
2
Department of Electrical Eng and Information Techn, Iranian Research Organization for Science and Tech
Abstract
In this paper a new method is introduced for root detection in minirhizotron images for root investigation. In this method firstly a hypothesis testing framework is defined to separate roots from background and noise. Then the correct roots are extracted by using an entropy-based geometric level set decision function. Performance of the proposed method is evaluated on real captured images in two different scenarios. In the first scenario images contain several roots however the second scenario belongs to no-root images, which increases the chance of false detection error. The obtained results show the greater ability of the proposed method in root detection compared with present approaches in all examined images. Furthermore it can be shown that better detection of roots in proposed algorithm not only doesn't lead to extracting more false particles but also it decreases rate of false detections compared to existing algorithms.
Heidari, M., & shojaedini, S. V. (2014). A New Method for Root Detection in Minirhizotron Images: Hypothesis Testing Based on Entropy-Based Geometric Level Set Decision. International Journal of Engineering, 27(1), 91-100.
MLA
Masuod Heidari; seyed vahab shojaedini. "A New Method for Root Detection in Minirhizotron Images: Hypothesis Testing Based on Entropy-Based Geometric Level Set Decision". International Journal of Engineering, 27, 1, 2014, 91-100.
HARVARD
Heidari, M., shojaedini, S. V. (2014). 'A New Method for Root Detection in Minirhizotron Images: Hypothesis Testing Based on Entropy-Based Geometric Level Set Decision', International Journal of Engineering, 27(1), pp. 91-100.
VANCOUVER
Heidari, M., shojaedini, S. V. A New Method for Root Detection in Minirhizotron Images: Hypothesis Testing Based on Entropy-Based Geometric Level Set Decision. International Journal of Engineering, 2014; 27(1): 91-100.