Datasets
Standard Dataset
Carnolian Grey Honeybees dataset
- Citation Author(s):
- Submitted by:
- Ratko Pilipovic
- Last updated:
- Fri, 05/21/2021 - 02:36
- DOI:
- 10.21227/42s6-9764
- License:
136 Views- Categories:
Abstract
The images containing honey bees were extracted from the video recorded in the Botanic Garden of the University of Ljubljana, where a beehive with a colony of the Carnolian Grey, the native Slovene species, is placed. We set the camera above the beehive entrance and recorded the honey bees on the shelf in front of the entrance and the honey bees entering and exiting the hive. With such a setup, we ensured a non-invasive recording of the honey bees in their natural environment. The dataset contains 65 images of size 2688 x 1504 pixels. There is a total of 1040 ground-truth bounding boxes containing Carnolian Grey honey bees.
Cite
This data is provided here for research purpose(s) only. You are allowed to use this data provided that you cite the following papers:
Bulić, P.; Pilipović, R.; Risojević, V.; Lotrič, U. "An approximate GEMM unit for energy-efficient processing in convolutional neural networks." submitted to Sensors 2021
Data organization
- img/ -> Original images
- img_scale -> Scaled images for YOLO network (416,416)
- tags/bee_tags.txt -> tags for images in the img/ folder
- tags/bee_scaled_tags.txt -> tags for images in the img_scaled/ folder
Tags structure
Every line in the tag file contains the bounding boxes and labels of objects in a particular image. The lines have the following format:
path/to/image1 x01,y01,x02,y02,label0 x11,y11,x12,y12,label1 ..... xn1,yn1,xn2,yn2,labeln .......
Where (xn1,yn1) and (xn2,yn2) represents coordinates of the upper left corner and lower right corner of the n-th bounding box for image (path/to/image1), respectively. The (labeln) represents the bounding box label. As we differentiate between the background and honey bees, all labels are set to zero.
