|
|
Genovesi, B., Berrebi, P., Nagai, S., Reynaud, N., Wang, J., & Masseret, E. (2015). Geographic structure evidenced in the toxic dinoflagellate Alexandrium pacificum Litaker (A. catenella – group IV (Whedon & Kofoid) Balech) along Japanese and Chinese coastal waters. Marine Pollution Bulletin, 98(1–2), 95–105.
Résumé: Abstract
The intra-specific diversity and genetic structure within the Alexandrium pacificum Litaker (A. catenella – Group IV) populations along the Temperate Asian coasts, were studied among individuals isolated from Japan to China. The UPGMA dendrogram and FCA revealed the existence of 3 clusters. Assignment analysis suggested the occurrence of gene flows between the Japanese Pacific coast (cluster-1) and the Chinese Zhejiang coast (cluster-2). Human transportations are suspected to explain the lack of genetic difference between several pairs of distant Japanese samples, hardly explained by a natural dispersal mechanism. The genetic isolation of the population established in the Sea of Japan (cluster-3) suggested the existence of a strong ecological and geographical barrier. Along the Pacific coasts, the South–North current allows limited exchanges between Chinese and Japanese populations. The relationships between Temperate Asian and Mediterranean individuals suggested different scenario of large-scale dispersal mechanisms.
|
|
|
|
Villon, S., Mouillot, D., Chaumont, M., Darling, E. S., Subsol, G., Claverie, T., et al. (2018). A Deep learning method for accurate and fast identification of coral reef fishes in underwater images. Ecol. Inform., 48, 238–244.
Résumé: Identifying and counting fish individuals on photos and videos is a crucial task to cost-effectively monitor marine biodiversity, yet it remains difficult and time-consuming. In this paper, we present a method to assist the identification of fish species on underwater images, and we compare our model performances to human ability in terms of speed and accuracy. We first tested the performance of a convolutional neural network (CNN) trained with different photographic databases while accounting for different post-processing decision rules to identify 20 fish species. Finally, we compared the performance of species identification of our best CNN model with that of humans on a test database of 1197 fish images representing nine species. The best CNN was the one trained with 900,000 images including (i) whole fish bodies, (ii) partial fish bodies and (iii) the environment (e.g. reef bottom or water). The rate of correct identification was 94.9%, greater than the rate of correct identification by humans (89.3%). The CNN was also able to identify fish individuals partially hidden behind corals or behind other fish and was more effective than humans to identify fish on smallest or blurry images while humans were better to identify fish individuals in unusual positions (e.g. twisted body). On average, each identification by our best CNN using a common hardware took 0.06 s. Deep Learning methods can thus perform efficient fish identification on underwater images and offer promises to build-up new video-based protocols for monitoring fish biodiversity cheaply and effectively.
|
|
