Tobacco Dataset for crop/weed classification

We have acquired a new tobacco-weed dataset using a Mavic Mini drone. Eight fields of tobacco crop are captured in Mardan, Khyber Pakhtunkhwa, Pakistan. At different growth stages these eight fields are captured at crop age of 15 to 40 days approximately. Data is captured at 1920×1080-pixel resolution. Dataset is captured at an average altitude of 4 meters with ground sampling distance of 0.1 cm/pixel. Find Details in attached Research papers Citation Request: if you use these datasets in your research or projects by any means, please cite following publications. 1) Patch-wise weeds coarse segmentation mask from aerial imagery of sesame crop (Published in Computers and Electronics in Agriculture 2022, HEC Recognized W category, Impact factor 6.757, Q1) 2) Towards automated weed detection through two-stage semantic segmentation of tobacco and weed pixels in aerial Imagery (Published in Smart Agricultural Technology (A companion journal of Computers and Electronics in Agriculture)) 3) A Patch-Image Based Classification Approach for Detection of Weeds in Sugar Beet Crop (Published in IEEE Access, Impact factor 3.1, Q1) Acknowledgement Request This work is funded by the Higher Education Commission of Pakistan and the National center for Robotics and Automation (DF-1009–31). Please Acknowledge. Steps to Access Mendeley datasets 1. Click on the link 2. The link with ask you to sign in or register with institutional email. 3. Use your institutional/organization email to register and then sign in. 4. Once sign in, dataset will be visible in compressed folders 5. Download and unzip/umcompress folder 6. Use dataset in your research as you see fit (folders contains original images, and their labeled groundtruths, along with binary vegetation masks. In groundtruths background have label value of 0, crop have label 1 and weeds have label of 2. maskref subfolders shows labelled data for visualization) Find More datasets and published articles in Related Links

本研究采用大疆御Mini（Mavic Mini）无人机采集了一套全新的烟草-杂草数据集。数据集采集自巴基斯坦开伯尔-普赫图赫瓦省马尔丹地区的8块烟草田。上述8块田的影像采集于烟草生长周期15至40天左右的不同生育阶段。影像分辨率为1920×1080像素，采集时平均飞行高度为4米，地面采样距离（ground sampling distance）为0.1厘米/像素。

详见附随研究论文。引用说明：若您在任何研究或项目中使用本数据集，请引用以下文献：
1. 《基于芝麻航拍影像的杂草逐块粗分割掩码》（发表于《Computers and Electronics in Agriculture》2022年，获巴基斯坦高等教育委员会（Higher Education Commission, HEC）认定为W类期刊，影响因子6.757，属于Q1分区）
2. 《基于航拍影像两阶段语义分割实现烟草与杂草像素自动识别》（发表于《Smart Agricultural Technology》，该刊为《Computers and Electronics in Agriculture》的姊妹期刊）
3. 《基于块图像分类的甜菜田杂草检测方法》（发表于《IEEE Access》，影响因子3.1，Q1分区）

致谢说明：本研究受巴基斯坦高等教育委员会与国家机器人与自动化中心（National Center for Robotics and Automation, NCRA）资助（项目编号DF-1009–31），请在使用时予以致谢。

Mendeley数据集获取步骤：
1. 点击对应链接；
2. 链接将提示您使用机构邮箱登录或注册；
3. 使用所属机构/组织邮箱完成注册并登录；
4. 登录成功后，数据集将以压缩文件夹形式展示；
5. 下载并解压文件夹；
6. 可根据研究需求自由使用本数据集（文件夹内包含原始影像、标注真值图（groundtruths）以及二值植被掩码。真值图中，背景标签值为0，作物标签值为1，杂草标签值为2。maskref子文件夹用于展示可视化标注数据）。

更多数据集与已发表论文请见相关链接。

搜集汇总

背景与挑战

背景概述

该数据集是一个用于烟草作物与杂草分类的航空影像数据集，通过无人机在巴基斯坦的八个烟草田采集，覆盖作物生长期15至40天，包含高分辨率图像和详细的像素级标注。数据集提供原始图像、标注真值及植被掩码，总大小为7.37 GB，主要用于农业自动化中的杂草检测研究，并需机构邮箱注册访问。