A New Method for the Segmentation of Algae Images Using Non-Uniform Background Improvement and Support Vector Machine
| Autor: | Ezzatollah Salari, Kyle Dannemiller |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
0301 basic medicine
biology Computer science Image quality business.industry Feature extraction Pattern recognition Image segmentation biology.organism_classification Algal bloom Support vector machine 03 medical and health sciences Statistical classification 030104 developmental biology Algae Segmentation Artificial intelligence business |
| Zdroj: | EIT |
| DOI: | 10.1109/eit.2018.8500095 |
| Popis: | Algae growth is a natural occurrence in many areas including: freshwater lakes, ponds, gulfs and other bodies of water. The algae can benefit the environment they live in or damage it when a harmful algal bloom takes place. For this reason, the rapid and accurate classification of algae in micro-image samples taken from freshwater bodies becomes highly desirable before an actual bloom proliferates. This paper explores a new method designed to increase the quality of algae micro-images and its segmentation, thus improving two important steps involved in the automatic recognition and classification of algae in images. First, the algae image quality was enhanced through the use of a non-uniform background improvement method. This method enhances an image by adjusting the background to a chosen intensity. Then, the algae in the improved quality image is segmented from the background using a support vector machine. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|