Perrenial Plants Detection

Introduction #

Benedikt Geisler

The author of the Perrenial Plants Detection dataset explores automating weed removal in perennial plant cultivation, which currently involves physically demanding manual labor. They aim to answer two key questions: 1) Can optical RGB inspection combined with advanced machine learning reliably detect weeds? 2) Does providing a plant center keypoint in the dataset aid in automatic center identification?

Over seven weeks, images were collected at a southern German plant nursery using a Fuji X-T2 camera, resulting in 392 high-resolution images (24MP, 4000x6000px). The dataset includes sections for weed classification and keypoint detection, with some image overlap.

Expand

Homepage

Summary #

Perrenial Plants Detection is a dataset for an object detection task. Possible applications of the dataset could be in the agricultural industry.

The dataset consists of 392 images with 1159 labeled objects belonging to 14 different classes including not identified, weed, stellaria media, and other: veronica arvensis, cirsium arvense, poa anua, matricaria chamomilla, taraxacum officinale, sagina subulata, woods, cardamine hirsuta, elymus repens, portulaca oleracea, and artiplex oblongifolia.

Images in the Perrenial Plants Detection dataset have bounding box annotations. There are 2 (1% of the total) unlabeled images (i.e. without annotations). There are 3 splits in the dataset: train (294 images), val (77 images), and test (21 images). The dataset was released in 2021.

Here are the visualized examples for the classes:

Explore #

Perrenial Plants Detection dataset has 392 images. Click on one of the examples below or open "Explore" tool anytime you need to view dataset images with annotations. This tool has extended visualization capabilities like zoom, translation, objects table, custom filters and more. Hover the mouse over the images to hide or show annotations.

Sample annotation mask from Perrenial Plants Detection

Sample image from Perrenial Plants Detection

👀

Have a look at 392 images

View images along with annotations and tags, search and filter by various parameters

Class balance #

There are 14 annotation classes in the dataset. Find the general statistics and balances for every class in the table below. Click any row to preview images that have labels of the selected class. Sort by column to find the most rare or prevalent classes.

Rows 1-10 of 14

Class ㅤ	Images ㅤ	Objects ㅤ	Count on image average	Area on image average
not identified➔ rectangle	103	173	1.68	2.62%
weed➔ rectangle	99	262	2.65	0.57%
stellaria media➔ rectangle	90	189	2.1	4.51%
veronica arvensis➔ rectangle	71	130	1.83	4.94%
cirsium arvense➔ rectangle	70	77	1.1	5.99%
poa anua➔ rectangle	42	121	2.88	4.62%
matricaria chamomilla➔ rectangle	35	70	2	8.65%
taraxacum officinale➔ rectangle	16	18	1.12	5.72%
woods➔ rectangle	9	16	1.78	2.03%
sagina subulata➔ rectangle	9	91	10.11	1.39%

Co-occurrence matrix #

Co-occurrence matrix is an extremely valuable tool that shows you the images for every pair of classes: how many images have objects of both classes at the same time. If you click any cell, you will see those images. We added the tooltip with an explanation for every cell for your convenience, just hover the mouse over a cell to preview the description.

Images #

Explore every single image in the dataset with respect to the number of annotations of each class it has. Click a row to preview selected image. Sort by any column to find anomalies and edge cases. Use horizontal scroll if the table has many columns for a large number of classes in the dataset.

Object distribution #

Interactive heatmap chart for every class with object distribution shows how many images are in the dataset with a certain number of objects of a specific class. Users can click cell and see the list of all corresponding images.

Class sizes #

The table below gives various size properties of objects for every class. Click a row to see the image with annotations of the selected class. Sort columns to find classes with the smallest or largest objects or understand the size differences between classes.

Rows 1-10 of 14

Class	Object count	Avg area	Max area	Min area	Min height	Min height	Max height	Max height	Avg height	Avg height	Min width	Min width	Max width	Max width
weed rectangle	262	0.22%	3.58%	0%	26px	0.65%	854px	21.35%	180px	4.43%	26px	0.43%	786px	22.74%
stellaria media rectangle	189	2.15%	24.56%	0.04%	107px	2.67%	2370px	59.25%	583px	14.59%	84px	1.4%	2805px	46.75%
not identified rectangle	173	1.56%	9.14%	0.15%	151px	3.77%	1393px	34.83%	536px	13.39%	177px	2.95%	1670px	27.83%
veronica arvensis rectangle	130	2.71%	36.58%	0.05%	83px	2.08%	3579px	89.47%	632px	15.76%	144px	2.4%	2453px	40.88%
poa anua rectangle	121	1.61%	13.9%	0.12%	153px	3.83%	1662px	41.55%	549px	13.72%	146px	2.43%	2871px	47.85%
sagina subulata rectangle	91	0.14%	1.62%	0.01%	44px	1.1%	699px	17.48%	140px	3.49%	39px	0.65%	628px	10.47%
cirsium arvense rectangle	77	5.52%	52.69%	0.07%	161px	4.03%	2976px	86.11%	861px	21.76%	110px	1.83%	3172px	86.98%
matricaria chamomilla rectangle	70	4.34%	18.6%	0.24%	205px	5.12%	2343px	58.58%	948px	23.69%	270px	4.5%	1905px	31.75%
taraxacum officinale rectangle	18	5.09%	32.1%	0.04%	116px	2.9%	2342px	58.55%	801px	20.03%	89px	1.48%	3289px	54.82%
woods rectangle	16	1.14%	6.16%	0.03%	88px	2.2%	1181px	29.52%	403px	10.08%	77px	1.28%	1251px	20.85%

Spatial Heatmap #

The heatmaps below give the spatial distributions of all objects for every class. These visualizations provide insights into the most probable and rare object locations on the image. It helps analyze objects' placements in a dataset.

Objects #

Table contains all 1159 objects. Click a row to preview an image with annotations, and use search or pagination to navigate. Sort columns to find outliers in the dataset.

Rows 1-10 of 1159

Object ID ㅤ	Class ㅤ	Image name click row to open	Image size height x width	Height ㅤ	Height ㅤ	Width ㅤ	Width ㅤ	Area ㅤ
1➔	stellaria media rectangle	DSCF8556.JPG	4000 x 6000	288px	7.2%	322px	5.37%	0.39%
2➔	stellaria media rectangle	DSCF8556.JPG	4000 x 6000	814px	20.35%	1157px	19.28%	3.92%
3➔	cirsium arvense rectangle	DSCF8717.JPG	4000 x 6000	1055px	26.38%	776px	12.93%	3.41%
4➔	cirsium arvense rectangle	DSCF8822.JPG	4000 x 6000	1064px	26.6%	1159px	19.32%	5.14%
5➔	not identified rectangle	DSCF8595.JPG	4000 x 6000	322px	8.05%	409px	6.82%	0.55%
6➔	cirsium arvense rectangle	DSCF8646.JPG	4000 x 6000	609px	15.22%	596px	9.93%	1.51%
7➔	not identified rectangle	DSCF8592.JPG	4000 x 6000	330px	8.25%	513px	8.55%	0.71%
8➔	not identified rectangle	DSCF8592.JPG	4000 x 6000	297px	7.42%	326px	5.43%	0.4%
9➔	weed rectangle	DSCF8882.JPG	4000 x 6000	121px	3.02%	96px	1.6%	0.05%
10➔	weed rectangle	DSCF8882.JPG	4000 x 6000	54px	1.35%	51px	0.85%	0.01%

License #

Perrenial Plants Detection is under MIT license.

Citation #

If you make use of the Perrenial Plants Detection data, please cite the following reference:

@dataset{Perrenial Plants Detection,
  author={Benedikt Geisler},
  title={Perrenial Plants Detection},
  year={2021},
  url={https://www.kaggle.com/datasets/benediktgeisler/perrenial-plants-detection}
}

Source

If you are happy with Dataset Ninja and use provided visualizations and tools in your work, please cite us:

@misc{ visualization-tools-for-perrenial-plants-dataset,
  title = { Visualization Tools for Perrenial Plants Detection Dataset },
  type = { Computer Vision Tools },
  author = { Dataset Ninja },
  howpublished = { \url{ https://datasetninja.com/perrenial-plants } },
  url = { https://datasetninja.com/perrenial-plants },
  journal = { Dataset Ninja },
  publisher = { Dataset Ninja },
  year = { 2024 },
  month = { dec },
  note = { visited on 2024-12-15 },
}

Download #

Dataset Perrenial Plants Detection can be downloaded in Supervisely format:

As an alternative, it can be downloaded with dataset-tools package:

pip install --upgrade dataset-tools

… using following python code:

import dataset_tools as dtools

dtools.download(dataset='Perrenial Plants Detection', dst_dir='~/dataset-ninja/')

Make sure not to overlook the python code example available on the Supervisely Developer Portal. It will give you a clear idea of how to effortlessly work with the downloaded dataset.

The data in original format can be downloaded here.

. . .

Disclaimer #

Our gal from the legal dep told us we need to post this:

Dataset Ninja provides visualizations and statistics for some datasets that can be found online and can be downloaded by general audience. Dataset Ninja is not a dataset hosting platform and can only be used for informational purposes. The platform does not claim any rights for the original content, including images, videos, annotations and descriptions. Joint publishing is prohibited.

You take full responsibility when you use datasets presented at Dataset Ninja, as well as other information, including visualizations and statistics we provide. You are in charge of compliance with any dataset license and all other permissions. You are required to navigate datasets homepage and make sure that you can use it. In case of any questions, get in touch with us at hello@datasetninja.com.