Image segmentation evaluation: a survey of methods
- PDF / 2,304,940 Bytes
- 38 Pages / 439.37 x 666.142 pts Page_size
- 48 Downloads / 323 Views
Image segmentation evaluation: a survey of methods Zhaobin Wang1 · E. Wang1 · Ying Zhu2
© Springer Nature B.V. 2020
Abstract Image segmentation is a prerequisite for image processing. There are many methods for image segmentation, and as a result, a great number of methods for evaluating segmentation results have also been proposed. How to effectively evaluate the quality of image segmentation is very important. In this paper, the existing image segmentation quality evaluation methods are summarized, mainly including unsupervised methods and supervised methods. Based on hot issues, the application of metrics in natural, medical and remote sensing image evaluation is further outlined. In addition, an experimental comparison for some methods were carried out and the effectiveness of these methods was ranked. At the same time, the effectiveness of classical metrics for remote sensing and medical image evaluation is also verified. Keywords Image segmentation · Segmentation evaluation · Unsupervised evaluation · Supervised evaluation · Evaluation application
1 Introduction Image segmentation is a very important and difficult problem in many fields such as image processing, pattern recognition and artificial intelligence. The primary and important key step in computer vision technology is image segmentation, which is also an important part of image semantic understanding. Correct image processing is impossible without proper segmentation, so image segmentation is an important image analysis technique in different areas and it is applied widely, especially in medical image analysis (Domingo et al. 2016; Zhou et al. 2020; Kaya et al. 2017; Li et al. 2020; Goceri et al. 2015; Goceri and Songul 2017b, 2018; Goceri 2018) for anomaly detection, disease diagnosis or monitoring. Similarly, segmentation technology is also indispensable in remote sensing image processing (Wang et al. 2020; Zhang et al. 2020; Peng et al. 2019; Zeng et al. 2019; Nogueira et al. * Zhaobin Wang [email protected] * Ying Zhu [email protected] 1
School of Information Science and Engineering, Lanzhou University, Lanzhou 730000, China
2
Key Laboratory of Microbial Resources Exploitation and Application of Gansu Province, Institute of Biology, Gansu Academy of Sciences, Lanzhou, China
13
Vol.:(0123456789)
Z. Wang et al.
2019; Wu et al. 2019). In the research and application of images, people tend to be only interested in certain parts of the image (Göçeri 2013; Goceri 2016; Goceri and Songül 2017a; Goceri et al. 2017; Goceri 2019b). These parts are often referred to goals or prospects (other parts are called backgrounds). Goals or prospects generally refer to certain areas of an image with useful properties. In order to identify and analyze the targets in the image, these must be separated and extracted from the image. On this basis, the target can be further measured and the image can be further utilized. Image segmentation is the technique and process of dividing an image into regions with specific characteristics and extracti
Data Loading...
