RT info:eu-repo/semantics/article
T1 Lights and pitfalls of convolutional neural networks for diatom identification
A1 Pedraza Dorado, Aníbal
A1 Bueno García, María Gloria
A1 Déniz Suárez, Óscar
A1 Ruiz-Santaquiteria Alegre, Jesús
A1 Sánchez Bueno, Carlos
A1 Blanco Lanza, Saúl
A1 Borrego Ramos, María
A1 Olenici, Adriana
A1 Cristóbal Pérez, Gabriel
A2 Ecologia
K1 Ecología. Medio ambiente
K1 Deep learning
K1 CNN
K1 RCNN
K1 YOLO
K1 Diatoms detection
K1 Water quality
AB Diatom detection has been a challenging task for computer scientist and biologist during past years. In this work, the new state of art techniques based on the deep learning framework have been tested, in order to check whether they are suitable for this purpose. On the one hand, RCNNs (Region based Convolutional Neural Networks), which select candidate regions and applies a convolutional neural network and, on the other hand, YOLO (You Only Look Once), which applies a single neural network over the whole image, have been tested. The first one is able to reach poor results in out experimentation, with an average of 0.68 recall and some tricky aspects, as for example it is needed to apply a bounding box merging algorithm to get stable detections; but the second one gets remarkable results, with an average of 0.84 recall in the evaluation that have been carried out, and less aspects to take into account after the detection has been performed. Future work related to parameter tuning and processing are needed to increase the performance of deep learning in the detection task. However, as for classification it has been probed to provide succesfully performance.
PB SPIE
YR 2019
FD 2019-01-22
LK http://hdl.handle.net/10612/9312
UL http://hdl.handle.net/10612/9312
NO Anibal  Pedraza, Gloria  Bueno, Oscar  Déniz, Jesus  Ruiz-Santaquiteria,Carlos  Sanchez, Saul  Blanco, Maria  Borrego-Ramos, Adriana  Olenici,Gabriel  Cristobal, "Lights and pitfalls of convolutional neural networks fordiatom identification," Proc. SPIE 10679, Optics, Photonics, and DigitalTechnologies for Imaging Applications V, 106790G (24 May 2018): doi:10.1117/12.2309488
NO P. 1-10
DS BULERIA. Repositorio Institucional de la Universidad de León
RD 26-abr-2024