BrRacemeCounter: Mask R-CNN for Raceme Instance Segmentation

Bioversity International and International Center for Tropical Agriculture; Arrechea-Castillo, Darwin Alexis; Cardoso Arango, Juan Andres

BrRacemeCounter: Mask R-CNN for Raceme Instance Segmentation

Autores

Bioversity International and International Center for Tropical Agriculture

Arrechea-Castillo, Darwin Alexis

Cardoso Arango, Juan Andres

Descripción

This is an adapted implementation of Mask R-CNN on Python 3, Keras, and TensorFlow. The model generates bounding boxes and segmentation masks for each instance of an object in the image. It's based on Feature Pyramid Network (FPN) and a ResNet50 or ResNet101 backbone. This project detects and count the total amount of racemes in tropical forages (Urochloa spp.) throught an web or a desktop based application