LATIDIA · Robótica

Localización selectiva de bolos de algodón para la recolección robótica: evaluación de modelos de visión de aprendizaje profundo en condiciones de campo

arXiv: 2609.19592v1Tipo de anuncio: Cross Resumen: Este estudio desarrolló y evaluó un marco de percepción basado en el aprendizaje profundo para la recolección selectiva robótica de algodón. El conjunto de datos contenía 1.008 imágenes de campo anotadas col

La noticia

arXiv:2609.19592v1 Announce Type: cross Abstract: This study developed and evaluated a deep-learning-based perception framework for selective robotic cotton picking. The dataset contained 1,008 annotated field images collected using three cameras under varying natural lighting and weather conditions. Object-detection models from the YOLOv8 through YOLOv13 families were evaluated using their default configurations, while segmentation performance was assessed using YOLOv8-seg, YOLOv11-seg, YOLOv12-seg, the Segment Anything Model (SAM), SAMv2.1, FastSAM, and Grounded-SAM with the Recognize Anything Model (RAM). Among the detection models, GELAN-s achieved the most favorable balance between mean average precision (mAP) and inference speed, obtaining an mAP of 86.1%, precision of 81.6%, recall of 76.6%, and an F1-score of 79.0%, with