No abstract

Precision herbicide application can substantially reduce herbicide input and weed control cost in turfgrass management systems. Intelligent spot-spraying system predominantly relies on machine vision-based detectors for autonomous weed control. In this work, several deep convolutional neural networks (DCNN) were constructed for detection of dandelion (Taraxacum officinale Web.), ground ivy (Glechoma hederacea L.), and spotted spurge (Euphorbia maculata L.) growing in perennial ryegrass. When the networks were trained using a dataset containing a total of 15,486 negative (images contained perennial ryegrass with no target weeds) and 17,600 positive images (images contained target weeds), VGGNet achieved high F1 scores (≥0.9278), with high recall values (≥0.9952) for detection of E. maculata, G. hederacea, and T. officinale growing in perennial ryegrass. The F1 scores of AlexNet ranged from 0.8437 to 0.9418 and were generally lower than VGGNet at detecting E. maculata, G. hederacea, and T. officinale. GoogleNet is not an effective DCNN at detecting these weed species mainly due to the low precision values. DetectNet is an effective DCNN and achieved high F1 scores (≥0.9843) in the testing datasets for detection of T. officinale growing in perennial ryegrass. Moreover, VGGNet had the highest Matthews correlation coefficient (MCC) values, while GoogleNet had the lowest MCC values. Overall, the approach of training DCNN, particularly VGGNet and DetectNet, presents a clear path toward developing a machine vision-based decision system in smart sprayers for precision weed control in perennial ryegrass.

BACKGROUND: Weed infestations reduce turfgrass aesthetics and uniformity. Postemergence (POST) herbicides are applied uniformly on turfgrass, hence areas without weeds are also sprayed. Deep learning, particularly the architecture of convolutional neural network (CNN), is a state-of-art approach to recognition of images and objects. In this paper, we report deep learning CNN (DL-CNN) models that are remarkably accurate at detection of broadleaf weeds in turfgrasses. RESULTS: VGGNet was the best model for detection of various broadleaf weeds growing in dormant bermudagrass [Cynodon dactylon (L.)] andDetectNet was the best model for detection of cutleaf evening-primrose (Oenothera laciniata Hill) in bahiagrass (Paspalum notatum Flugge) when the learning rate policy was exponential decay. These models achieved high F 1 scores (>0.99) and overall accuracy (>0.99), with recall values of 1.00 in the testing datasets. CONCLUSION:The results of the present research demonstrate the potential for detection of broadleaf weed using DL-CNN models for detection of broadleaf weeds in turfgrass systems. Further research is required to evaluate weed control in field conditions using these models for in situ video input in conjunction with a smart sprayer.