SeSame / Weed Aerial Dataset
Description
Sesame-weed aerial dataset was recorded with the Phantom 3 standard drone and Agrocam NDVI. Photographs of sesame fields were taken near Ballo Shahabal, Jhang, Punjab, Pakistan. These campaigns range in length from 16 to 45 days for the sesame crop. A ground sampling distance of 0.33 cm/pixel is achieved by flying the drone at an average height of 15 feet. The three channels of the Agrocam camera—NIR, G, and B—when combined, produce NGB composite images. Agrocam lacks the R (red channel), therefore the green grass appears orange in the pictures. Pictures were captured for both datasets at a resolution of 1920 × 1080 pixels, however for faster processing, we cropped non-overlapping images to a size of 480 x 352 pixels. Using the MATLAB Image Labeler app, we labelled the photos by hand. A label image in 8-bit unsigned grayscale is produced by the software. The depicted image has the pixel values 0, 1, and 2 assigned to the background, crop, and weed, respectively. Since there are no aerial dataset available publicly to the best of our knowledge, these datasets could be utilized to support further research in Sesame crop. No. Field Attribute/ images Timing Around / Date Captured Soil Condition 1 S3 Test / 120 6:30am / 09 Aug 2020 Before irrigation 2 S1 Train / 600 8:30am / 09 Aug 2020 Before irrigation 3 S1 Test / 120 11:30am / 10 Aug 2020 Before irrigation 4 S1 Test / 120 6:00pm / 19 Aug 2020 Before irrigation 5 S4 Test / 120 8:30am / 21 Aug 2020 After irrigation 6 S2 Train / 600 2:00pm / 21 Aug 2020 After irrigation 7 S3 Test / 120 3:30pm / 28 Aug 2020 After irrigation 8 S4 Test / 120 6:00pm / 06 Sep 2020 After irrigation 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. Find More data and research papers in related links attached.
Files
Steps to reproduce
Sesame Aerial Datasets is the main folder, and it has two subfolders named training and testing. Training folder contains self-descriptive folders named Detected vegetation (contains 1080P images where background is removed), Patch images (Contains cropped 480*352 size images and corresponding labels), Binary vegetation mask (contains binary image which show background and vegetation), Original 1080P images (contains original images) and Labelled 1080P images (Contains labels of original 1080P images). Patch images folder contains three subfolders’ data (Cropped original 480*352 patch images), mask (corresponding 480*352 patch labels) and maskref (480*352 size labels just for visualization purpose). Testing folder contains six different campaigns data (named 1,2,4,5,7,8, campaign no 3 and 6 data are put inside training folder). Inside campaign folders there are self-descriptive folders. Contact Contributor in case of confusion Dr. Syed Imran Moazzam Shah +92-313-7233382 imoazzam@ceme.nust.edu.pk drimoazz@gmail.com