Online learning method based on support vector machine for metallographic image segmentation
- PDF / 2,412,283 Bytes
- 8 Pages / 595.276 x 790.866 pts Page_size
- 91 Downloads / 161 Views
ORIGINAL PAPER
Online learning method based on support vector machine for metallographic image segmentation Mingchun Li1
· Dali Chen1
· Shixin Liu1 · Dinghao Guo1
Received: 27 December 2019 / Revised: 5 September 2020 / Accepted: 7 September 2020 © Springer-Verlag London Ltd., part of Springer Nature 2020
Abstract The shape, size and distribution of the microstructure could certainly reveal mechanical properties. Therefore, it is important to segment the microstructure accurately. However, in view of the randomness in the metallographic detection period, a fixed segmentation algorithm with good performance in specific and known dataset may lead to wrong result when external condition changes. To solve the problem, we proposes an online learning pipeline to adjust the model adaptively by dynamic samples. First, a preprocessing method is deployed to smooth the uneven gray level. Next, local features and abstract features are extracted by a genetic algorithm-optimized boundary detector and a deep learning model, respectively. Then, online support vector machine is employed to update the model parameters under complex and changeable conditions in real time. Finally, a post-processing method is employed to get final result. A variety of experiments are presented to verify the effectiveness and the convergence of online algorithm. The experiment results indicate that the proposed pipeline can effectively extract local and abstract features, and the real-time updating model according to dynamic samples achieves state-of-the-art segmentation performance. Keywords Online learning · Support vector machine · Genetic algorithm · Metallographic image
1 Introduction Metallography is a discipline for revealing the details of microstructures [1]. A representative example for microstructure evaluation is to obtain the characterization of phase volume fraction in second-phase alloy systems by segmenting metallographic image. In fact, because of the variability of microstructure, uneven gray value and shadow, metallographic images are difficult to be segmented precisely [2]. To solve the problems, many scholars proposed their methods. Generally, the segmentation methods could be roughly divided into two categories according to the strategy they used: rule-based method and learning-based method. The paradigm of rule-based method is to segment objects based on preset rules. Specifically, some typical rulebased methods, such as minimum cross-entropy method [3],
B
Dali Chen [email protected] Mingchun Li [email protected]
1
College of Information Science and Engineering, Northeastern University, Shenyang 110819, China
Watershed-based algorithm [4], mean shift algorithm [5], directional wavelet transform [6] and simple non-iterative clustering (SNIC) method [7], are used for microstructure segmentation. In contrast, learning-based method does not specifically rely on preset rules. They make their own decisions by learning the extracted features. For metallographic image, feature space mostly includes gray value, texture, co
Data Loading...
