Deep convolutional neural networks for image-based Convolvulus sepium detection in sugar beet fields

This type of copy-paste approach has been successful in improving weed recognition in a couple of different papers in sugar beet/bindweed: https://t.co/RJAu9YJK1J and a similar approach in cotton/morningglory/grass: https://t.co/hZzHjNQlya

A recent article in Plant Methods looked at synthetic images for data augmentation in sugar beet: https://t.co/RJAu9Z1kTh The combo of field + synthetic data improved hedge bindweed detection from an accuracy of 0.59 -> 0.76, so it will be interestin

Deep convolutional neural networks for image-based Convolvulus sepium detection in sugar beet fields https://t.co/kDBB9TBsgQ