PT Unknown AU Juan A. Carvajal Ayala Dennis Romero Angel Sappa TI Fine-tuning based deep convolutional networks for lepidopterous genus recognition BT 21st Ibero American Congress on Pattern Recognition PY 2016 BP 467 EP 475 DI 10.1007/978-3-319-52277-7_57 AB This paper describes an image classification approach oriented to identify specimens of lepidopterous insects at Ecuadorian ecological reserves. This work seeks to contribute to studies in the area of biology about genus of butterflies and also to facilitate the registration of unrecognized specimens. The proposed approach is based on the fine-tuning of three widely used pre-trained Convolutional Neural Networks (CNNs). This strategy is intended to overcome the reduced number of labeled images. Experimental results with a dataset labeled by expert biologists is presented, reaching a recognition accuracy above 92%. ER