This paper describes the participation of the ECOUAN team in the LifeCLEF 2015 challenge. We used a deep learning approach in which the complete system was learned without hand-engineered components. We pre-trained a convolutional neural network using 1.8 million images and used a fine-tuning strategy to transfer learned recognition capabilities from general domains to the specific challenge of Plant Identification task. The classification accuracy obtained by our method outperformed the best result obtained in 2014. Our group obtained the 4th position among all teams and the 10th position among 18 runs.