Comparative Study on Metaheuristic-Based Feature Selection for Cotton Foreign Fibers Recognition
The excellent feature set or feature combination of cotton foreign fibers is great significant to improve the performance of machine-vision-based recognition system of cotton foreign fibers. To find the excellent feature sets of foreign fibers, in this pa
- PDF / 938,746 Bytes
- 11 Pages / 439.37 x 666.142 pts Page_size
- 58 Downloads / 141 Views
School of Digital Media, Shenzhen Institute of Information Technology, Shenzhen 518172, China [email protected], [email protected] 2 Key Laboratory of Symbolic Computation and Knowledge Engineer (Jilin University), Ministry of Education, Changchun 130012, Jilin, China 3 College of Information and Electrical Engineering, China Agricultural University, Beijing 100083, China [email protected] 4 College of Physics and Electronic Information, Wenzhou University, Wenzhou 325035, China [email protected] 5 College of Information, Guangdong Ocean University, Zhanjiang 524025, Guangdong, China [email protected]
Abstract. The excellent feature set or feature combination of cotton foreign fibers is great significant to improve the performance of machine-vision-based recognition system of cotton foreign fibers. To find the excellent feature sets of foreign fibers, in this paper presents three metaheuristic-based feature selection approaches for cotton foreign fibers recognition, which are particle swarm optimization, ant colony optimization and genetic algorithm, respectively. The k-nearest neighbor classifier and support vector machine classifier with k-fold cross validation are used to evaluate the quality of feature subset and identify the cotton foreign fibers. The results show that the metaheuristic-based feature selection methods can efficiently find the optimal feature sets consisting of a few features. It is highly significant to improve the performance of recognition system for cotton foreign fibers. Keywords: Metaheuristic system
Feature selection
Foreign fibers
Recognition
1 Introduction The cotton foreign fibers, such as ropes, wrappers, plastic films and so on, are closely related to the quality of the final cotton textile products [1]. In the recent years, the machine-vision-based recognition systems have been widely used to assess the quality of cottons [2, 3], in which classification accuracy is an key measure to validate the © IFIP International Federation for Information Processing 2016 Published by Springer International Publishing AG 2016. All Rights Reserved D. Li and Z. Li (Eds.): CCTA 2015, Part I, IFIP AICT 478, pp. 8–18, 2016. DOI: 10.1007/978-3-319-48357-3_2
Comparative Study on Metaheuristic-Based Feature Selection
9
performance of recognition systems. To improve the classification accuracy, finding the optimal feature sets with high accuracy is an efficient way due to because it can improve the accuracy and speed of recognition systems. Feature selection (FS) is a main approach to find the optimal feature sets by reduce the irrelevant or redundant features. Currently, FS has been used to the area of machine learning and data mining [4]. Since to find the optimum feature sets is a NP problem, the researchers begin to turn to find the near optimal feature set and have proposed many algorithms [5, 6]. Currently, metaheuristic algorithms have attracted so much attention, the representive algorithms are particle swarm optimization (PSO for short), ant colony optimization (ACO for short) and genetic algorithm
Data Loading...
