Evaluation of different deep convolutional neural networks for detection of broadleaf weed seedlings in wheat

Evaluation of two deep learning-based approaches for detecting weeds growing in cabbage

BACKGROUND

Machine vision-based precision weed management is a promising solution to substantially reduce herbicide input and weed control cost. The objective of this research was to compare two different deep learning-based approaches for detecting weeds in cabbage: (1) detecting weeds directly, and (2) detecting crops by generating the bounding boxes covering the crops and any green pixels outside the bounding boxes were deemed as weeds.

RESULTS

The precision, recall, F1-score, mAP0.5, mAP0.5:0.95 of You Only Look Once (YOLO) v5 for detecting cabbage were 0.986, 0.979, 0.982, 0.995, and 0.851, respectively, while these metrics were 0.973, 0.985, 0.979, 0.993, and 0.906 for YOLOv8, respectively. However, none of these metrics exceeded 0.891 when detecting weeds. The reduced performances for directly detecting weeds could be attributed to the diverse weed species at varying densities and growth stages with different plant morphologies. A segmentation procedure demonstrated its effectiveness for extracting weeds outside the bounding boxes covering the crops, and thereby realizing effective indirect weed detection.

CONCLUSION

The indirect weed detection approach demands less manpower as the need for constructing a large training dataset containing a variety of weed species is unnecessary. However, in a certain case, weeds are likely to remain undetected due to their growth in close proximity with crops and being situated within the predicted bounding boxes that encompass the crops. The models generated in this research can be used in conjunction with the machine vision subsystem of a smart sprayer or mechanical weeder. © 2024 Society of Chemical Industry.

Convolutional neural network based on the fusion of image classification and segmentation module for weed detection in alfalfa

BACKGROUND

Accurate and reliable weed detection in real time is essential for realizing autonomous precision herbicide application. The objective of this research was to propose a novel neural network architecture to improve the detection accuracy for broadleaf weeds growing in alfalfa.

RESULTS

A novel neural network, ResNet-101-segmentation, was developed by fusing an image classification and segmentation module with the backbone selected from ResNet-101. Compared with existing neural networks (AlexNet, GoogLeNet, VGG16, and ResNet-101), ResNet-101-segmentation improved the detection of Carolina geranium, catchweed bedstraw, mugwort and speedwell from 78.27% to 98.17%, from 79.49% to 98.28%, from 67.03% to 96.23%, and from 75.95% to 98.06%, respectively. The novel network exhibited high values of confusion matrices (>90%) when trained with sufficient data sets.

CONCLUSION

ResNet-101-segmentation demonstrated excellent performance compared with existing models (AlexNet, GoogLeNet, VGG16, and ResNet-101) for detecting broadleaf weeds growing in alfalfa. This approach offers a promising solution to increase the accuracy of weed detection, especially in cases where weeds and crops have similar plant morphology. © 2024 Society of Chemical Industry.

Detection and coverage estimation of purple nutsedge in turf with image classification neural networks

BACKGROUND

Accurate detection of weeds and estimation of their coverage is crucial for implementing precision herbicide applications. Deep learning (DL) techniques are typically used for weed detection and coverage estimation by analyzing information at the pixel or individual plant level, which requires a substantial amount of annotated data for training. This study aims to evaluate the effectiveness of using image-classification neural networks (NNs) for detecting and estimating weed coverage in bermudagrass turf.

RESULTS

Weed-detection NNs, including DenseNet, GoogLeNet and ResNet, exhibited high overall accuracy and F1 scores (≥0.971) throughout the k-fold cross-validation. DenseNet outperformed GoogLeNet and ResNet with the highest overall accuracy and F1 scores (0.977). Among the evaluated NNs, DenseNet showed the highest overall accuracy and F1 scores (0.996) in the validation and testing data sets for estimating weed coverage. The inference speed of ResNet was similar to that of GoogLeNet but noticeably faster than DenseNet. ResNet was the most efficient and accurate deep convolution neural network for weed detection and coverage estimation.

CONCLUSION

These results demonstrated that the developed NNs could effectively detect weeds and estimate their coverage in bermudagrass turf, allowing calculation of the herbicide requirements for variable-rate herbicide applications. The proposed method can be employed in a machine vision-based autonomous site-specific spraying system of smart sprayers. © 2024 Society of Chemical Industry.

Towards deep learning based smart farming for intelligent weeds management in crops

Introduction

Deep learning (DL) is a core constituent for building an object detection system and provides a variety of algorithms to be used in a variety of applications. In agriculture, weed management is one of the major concerns, weed detection systems could be of great help to improve production. In this work, we have proposed a DL-based weed detection model that can efficiently be used for effective weed management in crops.

Methods

Our proposed model uses Convolutional Neural Network based object detection system You Only Look Once (YOLO) for training and prediction. The collected dataset contains RGB images of four different weed species named Grass, Creeping Thistle, Bindweed, and California poppy. This dataset is manipulated by applying LAB (Lightness A and B) and HSV (Hue, Saturation, Value) image transformation techniques and then trained on four YOLO models (v3, v3-tiny, v4, v4-tiny).

Results and discussion

The effects of image transformation are analyzed, and it is deduced that the model performance is not much affected by this transformation. Inferencing results obtained by making a comparison of correctly predicted weeds are quite promising, among all models implemented in this work, the YOLOv4 model has achieved the highest accuracy. It has correctly predicted 98.88% weeds with an average loss of 1.8 and 73.1% mean average precision value.

Future work

In the future, we plan to integrate this model in a variable rate sprayer for precise weed management in real time.

WeedNet-R: a sugar beet field weed detection algorithm based on enhanced RetinaNet and context semantic fusion

Accurate and dependable weed detection technology is a prerequisite for weed control robots to do autonomous weeding. Due to the complexity of the farmland environment and the resemblance between crops and weeds, detecting weeds in the field under natural settings is a difficult task. Existing deep learning-based weed detection approaches often suffer from issues such as monotonous detection scene, lack of picture samples and location information for detected items, low detection accuracy, etc. as compared to conventional weed detection methods. To address these issues, WeedNet-R, a vision-based network for weed identification and localization in sugar beet fields, is proposed. WeedNet-R adds numerous context modules to RetinaNet's neck in order to combine context information from many feature maps and so expand the effective receptive fields of the entire network. During model training, meantime, a learning rate adjustment method combining an untuned exponential warmup schedule and cosine annealing technique is implemented. As a result, the suggested method for weed detection is more accurate without requiring a considerable increase in model parameters. The WeedNet-R was trained and assessed using the OD-SugarBeets dataset, which is enhanced by manually adding the bounding box labels based on the publicly available agricultural dataset, i.e. SugarBeet2016. Compared to the original RetinaNet, the mAP of the proposed WeedNet-R increased in the weed detection job in sugar beet fields by 4.65% to 92.30%. WeedNet-R's average precision for weed and sugar beet is 85.70% and 98.89%, respectively. WeedNet-R outperforms other sophisticated object detection algorithms in terms of detection accuracy while matching other single-stage detectors in terms of detection speed.

Evaluation of convolutional neural networks for herbicide susceptibility-based weed detection in turf

Deep learning methods for weed detection typically focus on distinguishing weed species, but a variety of weed species with comparable plant morphological characteristics may be found in turfgrass. Thus, it is difficult for deep learning models to detect and distinguish every weed species with high accuracy. Training convolutional neural networks for detecting weeds susceptible to herbicides can offer a new strategy for implementing site-specific weed detection in turf. DenseNet, EfficientNet-v2, and ResNet showed high F₁ scores (≥0.986) and MCC values (≥0.984) to detect and distinguish the sub-images containing dollarweed, goosegrass, old world diamond-flower, purple nutsedge, or Virginia buttonweed growing in bermudagrass turf. However, they failed to reliably detect crabgrass and tropical signalgrass due to the similarity in plant morphology. When training the convolutional neural networks for detecting and distinguishing the sub-images containing weeds susceptible to ACCase-inhibitors, weeds susceptible to ALS-inhibitors, or weeds susceptible to synthetic auxin herbicides, all neural networks evaluated in this study achieved excellent F₁ scores (≥0.995) and MCC values (≥0.994) in the validation and testing datasets. ResNet demonstrated the fastest inference rate and outperformed the other convolutional neural networks on detection efficiency, while the slow inference of EfficientNet-v2 may limit its potential applications. Grouping different weed species growing in turf according to their susceptibility to herbicides and detecting and distinguishing weeds by herbicide categories enables the implementation of herbicide susceptibility-based precision herbicide application. We conclude that the proposed method is an effective strategy for site-specific weed detection in turf, which can be employed in a smart sprayer to achieve precision herbicide spraying.

A deep learning-based method for classification, detection, and localization of weeds in turfgrass

BACKGROUND

Precision spraying of synthetic herbicides can reduce herbicide input. Previous research demonstrated the effectiveness of using image classification neural networks for detecting weeds growing in turfgrass, but did not attempt to discriminate weed species and locate the weeds on the input images. The objectives of this research were to: (i) investigate the feasibility of training deep learning models using grid cells (subimages) to detect the location of weeds on the image by identifying whether or not the grid cells contain weeds; and (ii) evaluate DenseNet, EfficientNetV2, ResNet, RegNet and VGGNet to detect and discriminate multiple weed species growing in turfgrass (multi-classifier) and detect and discriminate weeds (regardless of weed species) and turfgrass (two-classifier).

RESULTS

The VGGNet multi-classifier exhibited an F₁ score of 0.950 when used to detect common dandelion and achieved high F₁ scores of ≥0.983 to detect and discriminate the subimages containing dallisgrass, purple nutsedge and white clover growing in bermudagrass turf. DenseNet, EfficientNetV2 and RegNet multi-classifiers exhibited high F₁ scores of ≥0.984 for detecting dallisgrass and purple nutsedge. Among the evaluated neural networks, EfficientNetV2 two-classifier exhibited the highest F₁ scores (≥0.981) for exclusively detecting and discriminating subimages containing weeds and turfgrass.

CONCLUSION

The proposed method can accurately identify the grid cells containing weeds and thus precisely locate the weeds on the input images. Overall, we conclude that the proposed method can be used in the machine vision subsystem of smart sprayers to locate weeds and make the decision for precision spraying herbicides onto the individual map cells. © 2022 Society of Chemical Industry.

Deep learning for detecting herbicide weed control spectrum in turfgrass

BACKGROUND

Precision spraying of postemergence herbicides according to the herbicide weed control spectrum can substantially reduce herbicide input. The objective of this research was to evaluate the effectiveness of using deep convolutional neural networks (DCNNs) for detecting and discriminating weeds growing in turfgrass based on their susceptibility to ACCase-inhibiting and synthetic auxin herbicides.

RESULTS

GoogLeNet, MobileNet-v3, ShuffleNet-v2, and VGGNet were trained to discriminate the vegetation into three categories based on the herbicide weed control spectrum: weeds susceptible to ACCase-inhibiting herbicides, weeds susceptible to synthetic auxin herbicides, and turfgrass without weed infestation (no herbicide). ShuffleNet-v2 and VGGNet showed high overall accuracy (≥ 0.999) and F₁ scores (≥ 0.998) in the validation and testing datasets to detect and discriminate weeds susceptible to ACCase-inhibiting and synthetic auxin herbicides. The inference time of ShuffleNet-v2 was similar to MobileNet-v3, but noticeably faster than GoogLeNet and VGGNet. ShuffleNet-v2 was the most efficient and reliable model among the neural networks evaluated.

CONCLUSION

These results demonstrated that the DCNNs trained based on the herbicide weed control spectrum could detect and discriminate weeds based on their susceptibility to selective herbicides, allowing the precision spraying of particular herbicides to susceptible weeds and thereby saving more herbicides. The proposed method can be used in a machine vision-based autonomous spot-spraying system of smart sprayers.

Automatic Recognition of Faults in Mining Areas Based on Convolutional Neural Network

Tectonic interpretation is critical to a coal mine’s safe production, and fault interpretation is an essential component of seismic tectonic interpretation. With the increasing necessity for accuracy in fault interpretation in coal mines, it is increasingly challenging to achieve greater accuracy only through traditional fault interpretation. The convolutional neural network (CNN) is a machine learning method established in recent years and it has been widely applied in coal mine fault interpretation because of its powerful feature-learning and classification capabilities. To improve the accuracy and efficiency of fault interpretation in coal mines, an automatic seismic fault identification method based on the convolutional neural network has been developed. Taking a mining area in eastern Yunnan province as an example, the CNN model realized automatic identification of faults with eight seismic attributes as feature inputs, and the model-training parameters were optimized and compared. Ten faults in the area were selected to analyze the prediction effect, and a comparative experiment was done with model structure parameters and training sets. The experimental results indicate that the training parameters have a significant influence on the training time and testing accuracy of the model, while structural parameters and training sets affect the actual prediction effect of the model. By comparison, the fault results predicted by the convolutional neural network are in good agreement with the manual interpretation, and the accuracy of the model is more than 85%, which proves that this method has certain feasibility and provides a new way to shorten the fault interpretation period and improve the interpretation accuracy.