A database of weed plants in the European part of Russia

Abstract

Weeds are plants that, although not specially cultivated, grow and often adapt to growing in arable lands. They form an ecological variant of flora, as a historically-formed set of species growing on cultivated soils. For the rational use of the chemical and biological crop protection products and to produce safe and high-quality food, up-to-date data on the floristic diversity of weeds and the patterns of its geographical change are required. The need for a weeds' database arises that allows many specialists to work together independently. However, the great value of any database lies not in its existence, but in the accumulation of data that can be used to analyse the factors affecting the species diversity of weeds. A dataset of weed species diversity and their distribution in the European part of Russia, based on the results of the authors' own research from 1999 to 2019, has been created. The dataset includes 24,284 observations of occurrences of weed plants, which were obtained on the basis of 2,049 relevés of segetal plant communities in seven regions of the European part of Russia. In total, the dataset includes information about 329 species of vascular plants growing in 65 farmlands: cereals, spring and winter crops, industrial crops, row crops and perennial grasses (Tretyakova et al. 2020).

Weed25: A deep learning dataset for weed identification

“…In terms of diversity, the largest dataset ( Tretyakova et al., 2020 ) was comprised of 329 categories, while the smallest dataset ( Jiang et al., 2018 ) had only four categories. Although the dataset created by Tretyakova et al. (2020) contained 329 categories, the images were not only weeds but also plants such as grains and economic crops.…”

“…The species average of this dataset was 73, which was usually not sufficient for model training. Moreover, most datasets were not open-access ( Jiang et al., 2018 ; Sudars et al., 2020 ; Tretyakova et al., 2020 ). Majority of the existing datasets had a certain imbalance on species diversity and evenness ( Giselsson et al., 2017 ; Sudars et al., 2020 ; Fan et al., 2021 ), which led to difficulties in practical applications.…”

“…To show the advantages of Weed25 in terms of species diversity and species average (species diversity: the number of all weed species in the dataset, abbreviated as diversity, that was characterized in this paper by the number of species; species average: the mean of the image number of each weed, abbreviated as diversity average, that was used in this paper to characterize the average number of weeds), Weed25 was compared with several existing datasets related to weed identification, as shown in Table 3. In terms of diversity, the largest dataset (Tretyakova et al, 2020) was comprised of 329 categories, while the smallest dataset (Jiang et al, 2018) (2020) contained 329 categories, the images were not only weeds but also plants such as grains and economic crops. The species average of this dataset was 73, which was usually not sufficient for model training.…”

“…On average, each category contained 80 images. Tretyakova et al (2020) sorted out a plant dataset containing 24,284 images of 329 plant species, with an average of 73 images for each category, which mainly incorporated grain, spring and winter crops, economic crops, and weeds, so it was not a purely weed dataset. Khan et al (2020) proposed a study of four publicly available datasets, including the Rice Sowing and Weed Dataset, the BoniRob Dataset, the Carrot Crop and Weed Dataset, and the Rice Millet Dataset.…”

“…On average, each category contained 80 images. Tretyakova et al. (2020) sorted out a plant dataset containing 24,284 images of 329 plant species, with an average of 73 images for each category, which mainly incorporated grain, spring and winter crops, economic crops, and weeds, so it was not a purely weed dataset.…”