The MegaWeeds dataset consists of seven existing datasets:

- WeedCrop dataset; Sudars, K., Jasko, J., Namatevs, I., Ozola, L., & Badaukis, N. (2020). Dataset of annotated food crops and weed images for robotic computer vision control. Data in Brief, 31, 105833. https://doi.org/https://doi.org/10.1016/j.dib.2020.105833

- Chicory dataset; Gallo, I., Rehman, A. U., Dehkord, R. H., Landro, N., La Grassa, R., & Boschetti, M. (2022). Weed detection by UAV 416a Image Dataset. https://universe.roboflow.com/chicory-crop-weeds-5m7vo/weed-detection-by-uav-416a/dataset/1

- Sesame dataset; Utsav, P., Raviraj, P., & Rayja, M. (2020). crop and weed detection data with bounding boxes. https://www.kaggle.com/datasets/ravirajsinh45/crop-and-weed-detection-data-with-bounding-boxes

- Sugar beet dataset; Wangyongkun. (2020). sugarbeetsAndweeds. https://www.kaggle.com/datasets/wangyongkun/sugarbeetsandweeds

- Weed-Detection-v2; Tandon, K. (2021, June). Weed_Detection_v2. https://www.kaggle.com/datasets/kushagratandon12/weed-detection-v2

- Maize dataset; Correa, J. M. L., D. Andújar, M. Todeschini, J. Karouta, JM Begochea, & Ribeiro A. (2021). WeedMaize. Zenodo. https://doi.org/10.5281/ZENODO.5106795

- CottonWeedDet12 dataset; Dang, F., Chen, D., Lu, Y., & Li, Z. (2023). YOLOWeeds: A novel benchmark of YOLO object detectors for multi-class weed detection in cotton production systems. Computers and Electronics in Agriculture, 205, 107655. https://doi.org/https://doi.org/10.1016/j.compag.2023.107655

All the datasets contain open-field images from crops and weeds with annotations. The annotation files were converted to text files so it can be used in the YOLO model. All the datasets were combined into one big dataset with in total 19,317 images. The dataset is split into a training and validation set.

WeedCrop Image Dataset

Data Description

It includes 2822 images. Weed are annotated in YOLO v5 PyTorch format.

The following pre-processing was applied to each image: * Auto-orientation of pixel data (with EXIF-orientation stripping)

The following augmentation was applied to create 3 versions of each source image: * Equal probability of one of the following 90-degree rotations: none, clockwise, counter-clockwise * Random shear of between -15° to +15° horizontally and -15° to +15° vertically * Random brigthness adjustment of between -25 and +25 percent

Classes

Crop, Weed

Inspiration

Identifying weeds and distinguish them from crops is very essential in Farming.

Acknowledgements

This dataset is derived by the following publication:

Kaspars Sudars, Janis Jasko, Ivars Namatevs, Liva Ozola, Niks Badaukis, Dataset of annotated food crops and weed images for robotic computer vision control, Data in Brief, Volume 31, 2020, 105833, ISSN 2352-3409, https://doi.org/10.1016/j.dib.2020.105833. (https://www.sciencedirect.com/science/article/pii/S2352340920307277) Abstract: Weed management technologies that can identify weeds and distinguish them from crops are in need of artificial intelligence solutions based on a computer vision approach, to enable the development of precisely targeted and autonomous robotic weed management systems. A prerequisite of such systems is to create robust and reliable object detection that can unambiguously distinguish weed from food crops. One of the essential steps towards precision agriculture is using annotated images to train convolutional neural networks to distinguish weed from food crops, which can be later followed using mechanical weed removal or selected spraying of herbicides. In this data paper, we propose an open-access dataset with manually annotated images for weed detection. The dataset is composed of 1118 images in which 6 food crops and 8 weed species are identified, altogether 7853 annotations were made in total. Three RGB digital cameras were used for image capturing: Intel RealSense D435, Canon EOS 800D, and Sony W800. The images were taken on food crops and weeds grown in controlled environment and field conditions at different growth stages Keywords: Computer vision; Object detection; Image annotation; Precision agriculture; Crop growth and development

Many thanks to Roboflow team for sharing this data.

Weed Detection Annotated Data

## Overview

Weed Detection Annotated Data is a dataset for instance segmentation tasks - it contains Weed Crops JTy0 annotations for 1,176 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

Weeds are invasive plants that compete with crops for vital nutrients and often attract pests, significantly impacting agricultural productivity. They account for approximately 45% of the annual productivity loss in farming. Manual weeding methods, while effective, are labor-intensive and financially burdensome, particularly for smallholder farmers. Conversely, excessive reliance on chemical herbicides has led to herbicide resistance in several weed species, creating additional challenges in weed management.

Emerging technologies, particularly artificial intelligence (AI) and computer vision, are revolutionizing agriculture by automating labor-intensive tasks. Computer vision enables the precise identification of crops and weeds from images, supporting autonomous systems for selective weeding and targeted herbicide application. To develop robust AI models, high-quality datasets are critical. Addressing this need, we introduce the MH-Weed16 Image Dataset, collected from soybean fields in the Maharashtra region of India between July 2023 and November 2023 under diverse natural field conditions. The dataset comprises a total of 18,677 images of 16 weed species, annotated with the guidance of agricultural experts. It also includes 7,577 representative crop samples, with 6,656 weed samples annotated using bounding boxes. Images of crop–weed interactions were captured from a top-down perspective to allow accurate weed area estimation based on bounding box annotations. Importantly, the dataset incorporates 282 UAV-captured images, providing a large-scale aerial perspective that complements ground-based close-range details. These UAV images enhance the dataset’s diversity by introducing varying resolutions, field scales, and occlusion conditions, making it suitable for both macro-level weed distribution mapping and micro-level species identification. This multi-platform inclusion strengthens the dataset’s applicability to precision agriculture, enabling the training and evaluation of advanced computer vision models for object detection, classification, and weed–crop discrimination under real-world field conditions.

Weed Detection

## Overview

Weed Detection is a dataset for object detection tasks - it contains Crop annotations for 1,297 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

This dataset is derived by the following publication:

Kaspars Sudars, Janis Jasko, Ivars Namatevs, Liva Ozola, Niks Badaukis, Dataset of annotated food crops and weed images for robotic computer vision control, Data in Brief, Volume 31, 2020, 105833, ISSN 2352-3409, https://doi.org/10.1016/j.dib.2020.105833. (https://www.sciencedirect.com/science/article/pii/S2352340920307277) Abstract: Weed management technologies that can identify weeds and distinguish them from crops are in need of artificial intelligence solutions based on a computer vision approach, to enable the development of precisely targeted and autonomous robotic weed management systems. A prerequisite of such systems is to create robust and reliable object detection that can unambiguously distinguish weed from food crops. One of the essential steps towards precision agriculture is using annotated images to train convolutional neural networks to distinguish weed from food crops, which can be later followed using mechanical weed removal or selected spraying of herbicides. In this data paper, we propose an open-access dataset with manually annotated images for weed detection. The dataset is composed of 1118 images in which 6 food crops and 8 weed species are identified, altogether 7853 annotations were made in total. Three RGB digital cameras were used for image capturing: Intel RealSense D435, Canon EOS 800D, and Sony W800. The images were taken on food crops and weeds grown in controlled environment and field conditions at different growth stages Keywords: Computer vision; Object detection; Image annotation; Precision agriculture; Crop growth and development

Weeds are undesired plants in agricultural fields that affect crop yield and quality by competing for nutrients, water, sunlight and space. Site-specific weed management (SSWM) through variable rate herbicide application and mechanical weed control have long been recommended in order to reduce the amount of herbicide and impact caused by uniform spraying. Accurate detection and classification of weeds in crop fields is a crucial first step for implementing such precise strategies. Drones are commonly used for image capturing but high wind pressure and different drone settings have a severe effect on the image quality, which potentially results in degraded images, e.g. due to motion blur. We publish a manually annotated and expert curated drone image dataset for weed detection in sorghum fields under challenging conditions. Our results show that our trained models generalize well regarding the detection of weeds, even for degraded captures due to motion blur. An UNet-like architecture with ResNet-34 as feature extractor achieved an F1-score of over 89 % on a hold-out test-set. Further analysis indicate that the trained model performed well in predicting the general plant shape, while most mis-classifications appeared at borders of the plants. Beyond that, our approach can detect intra-row weeds without additional information as well as partly occluded plants in contrast to existing research.

Github link: https://github.com/grimmlab/UAVWeedSegmentation

Please cite our original publication if you have used the data in your project or in any follow-up analysis (https://doi.org/10.1016/j.compag.2022.107388):

@article{GENZE2022107388, title = {Deep learning-based early weed segmentation using motion blurred UAV images of sorghum fields}, journal = {Computers and Electronics in Agriculture}, volume = {202}, pages = {107388}, year = {2022}, issn = {0168-1699}, doi = {https://doi.org/10.1016/j.compag.2022.107388}, url = {https://www.sciencedirect.com/science/article/pii/S0168169922006962}, author = {Nikita Genze and Raymond Ajekwe and Zeynep Güreli and Florian Haselbeck and Michael Grieb and Dominik G. Grimm}, keywords = {Deep learning, Weed detection, Weed segmentation, UAV, Precision agriculture}, }

The AgriAdapt dataset is a UAV-acquired aerial imagery collection designed for real-time weed detection in salad crops. It contains 643 high-quality RGB images, captured with a custom-built UAV platform at low altitude (5 m), yielding a ground sampling distance (GSD) of approximately 0.4 cm/pixel.

Images were collected across two fields in Rome, Italy, during different growth stages of salad crops (25–65 days). The dataset is divided into two partitions:

Field_ID_1: 322 annotated images acquired in September 2023 from three short flights over adjacent salad fields.

Field_ID_2: 321 annotated images acquired in October 2023 from a nearby larger field, representing a later growth stage.

Each image is annotated in YOLOv7 PyTorch format, distinguishing crops and weeds. Labels were manually created and quality-checked to ensure reliability. Minimal preprocessing was applied: images were resized to 640×640 pixels and EXIF orientation data was removed.

Figures below illustrate typical UAV field imagery and annotated masks, where weeds are highlighted in red for clarity.

This dataset fills a major gap in UAV-based agricultural datasets, as leafy vegetables and salad crops are underrepresented in existing public collections (which largely focus on crops such as sugar beet, cotton, maize, or soybean). The AgriAdapt dataset is therefore a valuable benchmark for researchers developing lightweight, real-time computer vision models deployable on embedded UAV hardware.

Key Features

Crop focus: Salad crops (leafy vegetables).

Image type: RGB aerial images (1280×1280 px, downsampled to 640×640).

Annotations: YOLOv7 object detection format (crop vs. weed).

Acquisition: Low-altitude UAV flights (5 m AGL, 0.4 cm/pixel GSD).

Dataset size: 643 images (322 in Field_ID_1, 321 in Field_ID_2).

Geographic location: Rome, Italy.

Conditions: Warm soil, moderate moisture, partly cloudy skies, strong sunlight.

Intended use: Weed detection, semantic segmentation, real-time inference on UAV-compatible hardware.

Weed Detection (Unsupervised + Supervised Learning)

Overview

This dataset is designed to support both supervised and unsupervised learning for the task of weed detection in crop fields. It provides labeled data in YOLO format suitable for training object detection models, unlabeled data for semi-supervised or unsupervised learning, and a separate test set for evaluation. The objective is to detect and distinguish between weed and crop instances using deep learning models like YOLOv5 or YOLOv8.

Dataset Structure

│ ├── labeled/ │ ├── images/ # Labeled images for training │ └── labels/ # YOLO-format annotations │ ├── unlabeled/ # Unlabeled images for unsupervised or semi-supervised learning │ └── test/ ├── images/ # Test images └── labels/ # Ground truth annotations in YOLO format

The dataset consists of drone images that were obtained for agricultural field monitoring to detect weeds and crops through computer vision and machine learning approaches. The images were obtained through high-resolution UAVs and annotated using the LabelImg and Roboflow tool. Each image has a corresponding YOLO annotation file that contains bounding box information and class IDs for detected objects. The dataset includes:

Original images in .jpg format with a resolution of 585 × 438 pixels.

Annotation files (.txt) corresponding to each image, following the YOLO format: class_id x_center y_center width height.

A classes.txt file listing the object categories used in labeling (e.g., Weed, Crop).

The dataset is intended for use in machine learning model development, particularly for precision agriculture, weed detection, and plant health monitoring. It can be directly used for training YOLOv7 and other object detection models.

This dataset consists of weed and crop images, featuring five distinct weed species—Kochia, Horseweed, Water Hemp, Ragweed, and Redroot Pigweed—and eight crop species: Black Bean, Canola, Corn, Lentil, Field Pea, Flax, Soybean, and Sugar Beet. The images are in RGB format and labeled using the YOLO annotation format. This data can be leveraged to develop AI models aimed at accurate weed and crop identification.

Weeds in agricultural farms are aggressive growers which compete for nutrition and other resources with the crop and reduce production. The increasing use of chemicals to control them has inadvertent consequences to the human health and the environment. In this work, a novel neural network training method combining semantic graphics for data annotation and an advanced encoder–decoder network for (a) automatic crop line detection and (b) weed (wild millet) detection in paddy fields is proposed. The detected crop lines act as a guiding line for an autonomous weeding robot for inter-row weeding, whereas the detection of weeds enables autonomous intra-row weeding. The proposed data annotation method, semantic graphics, is intuitive, and the desired targets can be annotated easily with minimal labor. Also, the proposed “extended skip network” is an improved deep convolutional encoder–decoder neural network for efficient learning of semantic graphics. Quantitative evaluations of the proposed method demonstrated an increment of 6.29% and 6.14% in mean intersection over union (mIoU), over the baseline network on the task of paddy line detection and wild millet detection, respectively. The proposed method also leads to a 3.56% increment in mIoU and a significantly higher recall compared to a popular bounding box-based object detection approach on the task of wild–millet detection.