Weed Classification from Natural Corn Field-Multi-Plant Images Based on Shallow and Deep Learning.

Crop and weed discrimination in natural field environments is still challenging for implementing automatic agricultural practices, such as weed control. Some weed control methods have been proposed. However, these methods are still restricted as they are implemented under controlled conditions. The development of a sound weed control system begins by recognizing the crop and the different weed plants presented in the field. In this work, a classification approach of Zea mays L. (Crop), narrow-leaf weeds (NLW), and broadleaf weeds (BLW) from multi-plant images are presented. Moreover, a large image dataset was generated. Images were captured in natural field conditions, in different locations, and growing stages of the plants. The extraction of regions of interest (ROI) is carried out employing connected component analysis (CCA), whereas the classification of ROIs is based on Convolutional Neural Networks (CNN) and compared with a shallow learning approach. To measure the classification performance of both methods, accuracy, precision, recall, and F1-score metrics were used. The best alternative for the weed classification task at early stages of growth and in natural corn field environments was the CNN-based approach, as indicated by the 97% accuracy value obtained.

Weed Classification Using Explainable Multi-Resolution Slot Attention.

In agriculture, explainable deep neural networks (DNNs) can be used to pinpoint the discriminative part of weeds for an imagery classification task, albeit at a low resolution, to control the weed population. This paper proposes the use of a multi-layer attention procedure based on a transformer combined with a fusion rule to present an interpretation of the DNN decision through a high-resolution attention map. The fusion rule is a weighted average method that is used to combine attention maps from different layers based on saliency. Attention maps with an explanation for why a weed is or is not classified as a certain class help agronomists to shape the high-resolution weed identification keys (WIK) that the model perceives. The model is trained and evaluated on two agricultural datasets that contain plants grown under different conditions: the Plant Seedlings Dataset (PSD) and the Open Plant Phenotyping Dataset (OPPD). The model represents attention maps with highlighted requirements and information about misclassification to enable cross-dataset evaluations. State-of-the-art comparisons represent classification developments after applying attention maps. Average accuracies of 95.42% and 96% are gained for the negative and positive explanations of the PSD test sets, respectively. In OPPD evaluations, accuracies of 97.78% and 97.83% are obtained for negative and positive explanations, respectively. The visual comparison between attention maps also shows high-resolution information.

Hyperspectral Classification of Cyperus esculentus Clones and Morphologically Similar Weeds.

Cyperus esculentus (yellow nutsedge) is one of the world's worst weeds as it can cause great damage to crops and crop production. To eradicate C. esculentus, early detection is key-a challenging task as it is often confused with other Cyperaceae and displays wide genetic variability. In this study, the objective was to classify C. esculentus clones and morphologically similar weeds. Hyperspectral reflectance between 500 and 800 nm was tested as a measure to discriminate between (I) C. esculentus and morphologically similar Cyperaceae weeds, and between (II) different clonal populations of C. esculentus using three classification models: random forest (RF), regularized logistic regression (RLR) and partial least squares-discriminant analysis (PLS-DA). RLR performed better than RF and PLS-DA, and was able to adequately classify the samples. The possibility of creating an affordable multispectral sensing tool, for precise in-field recognition of C. esculentus plants based on fewer spectral bands, was tested. Results of this study were compared against simulated results from a commercially available multispectral camera with four spectral bands. The model created with customized bands performed almost equally well as the original PLS-DA or RLR model, and much better than the model describing multispectral image data from a commercially available camera. These results open up the opportunity to develop a dedicated robust tool for C. esculentus recognition based on four spectral bands and an appropriate classification model.

Performances of the LBP Based Algorithm over CNN Models for Detecting Crops and Weeds with Similar Morphologies.

Weed invasions pose a threat to agricultural productivity. Weed recognition and detection play an important role in controlling weeds. The challenging problem of weed detection is how to discriminate between crops and weeds with a similar morphology under natural field conditions such as occlusion, varying lighting conditions, and different growth stages. In this paper, we evaluate a novel algorithm, filtered Local Binary Patterns with contour masks and coefficient k (k-FLBPCM), for discriminating between morphologically similar crops and weeds, which shows significant advantages, in both model size and accuracy, over state-of-the-art deep convolutional neural network (CNN) models such as VGG-16, VGG-19, ResNet-50 and InceptionV3. The experimental results on the "bccr-segset" dataset in the laboratory testbed setting show that the accuracy of CNN models with fine-tuned hyper-parameters is slightly higher than the k-FLBPCM method, while the accuracy of the k-FLBPCM algorithm is higher than the CNN models (except for VGG-16) for the more realistic "fieldtrip_can_weeds" dataset collected from real-world agricultural fields. However, the CNN models require a large amount of labelled samples for the training process. We conducted another experiment based on training with crop images at mature stages and testing at early stages. The k-FLBPCM method outperformed the state-of-the-art CNN models in recognizing small leaf shapes at early growth stages, with error rates an order of magnitude lower than CNN models for canola-radish (crop-weed) discrimination using a subset extracted from the "bccr-segset" dataset, and for the "mixed-plants" dataset. Moreover, the real-time weed-plant discrimination time attained with the k-FLBPCM algorithm is approximately 0.223 ms per image for the laboratory dataset and 0.346 ms per image for the field dataset, and this is an order of magnitude faster than that of CNN models.

Glyphosate-resistant and glyphosate-susceptible Palmer amaranth (Amaranthus palmeri S. Wats.): hyperspectral reflectance properties of plants and potential for classification.

BACKGROUND: Palmer amaranth (Amaranthus palmeri S. Wats.) is a troublesome agronomic weed in the southern United States, and several populations have evolved resistance to glyphosate. This paper reports on spectral signatures of glyphosate-resistant (GR) and glyphosate-sensitive (GS) plants, and explores the potential of using hyperspectral sensors to distinguish GR from GS plants. RESULTS: GS plants have higher light reflectance in the visible region and lower light reflectance in the infrared region of the spectrum compared with GR plants. The normalized reflectance spectrum of the GR and GS plants had best separability in the 400-500 nm, 650-690 nm, 730-740 nm and 800-900 nm spectral regions. Fourteen wavebands from within or near these four spectral regions provided a classification of unknown set of GR and GS plants, with a validation accuracy of 94% for greenhouse-grown plants and 96% for field-grown plants. CONCLUSIONS: GR and GS Palmer amaranth plants have unique hyperspectral reflectance properties, and there are four distinct regions of the spectrum that can separate the GR from GS plants. These results demonstrate that hyperspectral imaging has potential application to distinguish GR from GS Palmer amaranth plants (without a glyphosate treatment), with future implications for glyphosate resistance management. Published 2014. This article is a U.S. Government work and is in the public domain in the USA.