Una inmersión profunda en el
comportamiento animal
a través de la
visión por computadora
Energía y
ambiente
¿Qué son?
Son árboles que alcanzan un tamaño de 5–20 m de alto. Hojas profundamente lobadas, escabrosas a casi glabras y ásperas en la haz, aplicado-tomentosas en el envés, con 28–43 pares de nervios secundarios partiendo de los nervios primarios más largos; pecíolos hasta 7 dm de largo, uncinado-puberulentos.
from ultralytics import YOLO
from roboflow import Roboflow
rf = Roboflow(api_key="********************")
project = rf.workspace("fish-tracking-s61ch").project("fish-tracking-8malz")
version = project.version(3)
dataset = version.download("yolov11")
model = YOLO("yolo11n.pt")
augmentation_params = {
"hsv_h": 0.05,
"hsv_v": 0.3
}
model.train(data="/home/datamarindo/Fish-tracking-3/data.yaml",
epochs = 50,
imgsz = 640,
device = 0,
augment=True,
**augmentation_params )
model.val(data="/home/datamarindo/Fish-tracking-3/data.yaml", model="/home/datamarindo/runs/detect/train/weights/best.pt")
Validating runs/detect/train4/weights/best.pt...
Ultralytics 8.3.74 🚀 Python-3.10.16 torch-2.6.0+cu124 CUDA:0 (NVIDIA GeForce RTX 3060, 11940MiB)
YOLO11n summary (fused): 238 layers, 2,582,347 parameters, 0 gradients, 6.3 GFLOPs
Class Images Instances Box(P R mAP50 mAP50-95): 100%|██████████| 1/1 [00:00<00:00, 11.38it/s]
all 10 161 0.935 0.882 0.943 0.721
Speed: 0.1ms preprocess, 5.1ms inference, 0.0ms loss, 1.9ms postprocess per image
Peces
libres en el mar
['nematode', 'beer_glass', 'car_mirror', 'coffee_mug', 'face_powder', 'golf_ball', 'loupe', 'matchstick', 'Petri_dish', 'pick']
,["snail", "shark", "jellyfish", "axolotol", "dugong", "bubble", "electric_ray"]
Salud ecosistémica y
visión por computadora
Keith, H. & Vardon, Michael & Stein, John & Stein, Janet & Lindenmayer, David. (2017). Ecosystem accounts define explicit and spatial trade-offs for managing natural resources. Nature Ecology & Evolution. 1. 10.1038/s41559-017-0309-1.
Barbero-García, I., Kuschnerus, M., Vos, S., & Lindenbergh, R. (2023). Automatic detection of bulldozer-induced changes on a sandy beach from video using YOLO algorithm. International Journal of Applied Earth Observation and Geoinformation, 117, 103185. https://doi.org/10.1016/J.JAG.2023.103185
Proença, M. C. and Mendes, R. N. (2024) Beach Surveillance: A Contribution to Automation. Journal of Geoscience and Environment Protection, 12, 155-163. doi: 10.4236/gep.2024.1212010.
F i n
Inecol, 2025