A Review of Clustering Methods in Microorganism Image Analysis
Clustering plays a great role in microorganism image segmentation, feature extraction and classification, in all major application areas of microorganisms (medical, environmental, industrial, science and agriculture). Clustering methods are used for many
- PDF / 496,762 Bytes
- 13 Pages / 439.37 x 666.142 pts Page_size
- 13 Downloads / 279 Views
Abstract Clustering plays a great role in microorganism image segmentation, feature extraction and classification, in all major application areas of microorganisms (medical, environmental, industrial, science and agriculture). Clustering methods are used for many years in microorganism image processing because they are simple algorithms, easy to apply and efficient. Thus, in order to clarify the potential of different clustering techniques in different application domains of microorganisms, we survey related works from the 1990s till now, while pinning out the specific challenges on each work (area) with the corresponding suitable clustering algorithm. Keywords Microorganism image · Unsupervised learning · Image clustering · Feature extraction · Image segmentation
C. Li · F. Kulwa · J. Zhang · Z. Li · H. Xu Microscopic Image and Medical Image Analysis Group, MBIE College, Northeastern University, Shenyang, China e-mail: [email protected] F. Kulwa e-mail: [email protected] J. Zhang e-mail: [email protected] Z. Li e-mail: [email protected] H. Xu e-mail: [email protected] X. Zhao (B) Environmental Engineering, Northeastern University, Shenyang, China e-mail: [email protected] © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 E. Pi¸etka et al. (eds.), Information Technology in Biomedicine, Advances in Intelligent Systems and Computing 1186, https://doi.org/10.1007/978-3-030-49666-1_2
13
14
C. Li et al.
1 Introduction Microorganisms are very small living organisms which can appear as unicellular, multicellular and acellular types [20]. Some are beneficial and others are harmful to human [17]. Therefore, a depth understanding of microorganisms and their living habitat is of paramount importance, so as to leverage the beneficial ones. Microorganisms find roles in many areas such as agriculture [2], food [14], environments [34], industries [27], medical [12], water-borne [21] and science [7]. Figure 1. shows example of microorganisms in different application areas. In recent years, automatic image techniques have been used in identification, classification, segmentation and analysis of microorganisms, due to their advantage over traditional ways such as being consistent, accurate, fast and realiable [13, 19, 38]. Clustering methods plays an important role in microorganism image tracking, monitoring, segmentation, feature extraction and classification [8, 15]. Clustering methods are very useful in all microorganism application areas, because they are suitable in many image challeges such as uneven background noise on images [36]. Moreover, they are unsupervised or semi-supervised, and capable of capturing correlation between multiple objects in the image depending on subjected constraints (features) [9]. Clustering is the process of automatically classifying image dataset into a number of disjoint groups or clusters [37]. Because of the role and advantage of clustering methods in image (data) correlation and mining, a countable number
Data Loading...
