In this paper, we propose a method for counting motorcycles from images based on the selective search and deep learning. In the proposed approach, the objects in an image are segmented by the selective searching algorithm and then recognized individually by the Convolutional Neural Network (CNN). Finally, the objects recognized as motorcycles are counted. The experimental results show the proposed method is effective in counting the motorcycles.
CITATION STYLE
Hong, T. P., Yang, Y. C., & Su, J. H. (2019). Using selective search and CNN for counting motorcycles in images. In Smart Innovation, Systems and Technologies (Vol. 128, pp. 304–308). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-04585-2_37
Mendeley helps you to discover research relevant for your work.