What is it about?
Recognizing different animal species automatically, based on images, is a very challenging task from a computational point of view. But it is important for different activities, such as species preservation. One of the main tasks involved in this recognition is image segmentation, which corresponds to finding the regions of interest in the image. In this work, we developed an image segmentation approach that combines different image processing strategies in order to obtain a more precise and robust solution.
Featured Image
Photo by Kelsey Brown on Unsplash
Why is it important?
Bees are extremely important for the maintenance of life on earth. To help their preservation, the identification of bee species is needed. Computer vision and artificial intelligence techniques can be used to automate the process of species identification using images of the bees' wings. In this work, we develop and test an approach to segment bee wing images and extract features. The results indicate the developed approach is more accurate than related approaches.
Perspectives
The correct segmentation of bee wing images is a very important task for the automatic identification of species. This process, combined with an artificial intelligence classifier, can reach an accuracy of over 95% in species identification.
Dr. Luciano Digiampietri
Universidade de Sao Paulo Campus da Capital
Read the Original
This page is a summary of: Fully automatic segmentation of bee wing images, Revista Brasileira de Computação Aplicada, June 2020, UPF Editora,
DOI: 10.5335/rbca.v12i2.10420.
You can read the full text:
Contributors
The following have contributed to this page
