Classification of plant seedlings using deep convolutional neural network architectures

Weed management has a vital role in applications of agriculture domain. One of the key tasks is to identify the weeds after few days of plant germination which helps the farmers to perform early-stage weed management to reduce the contrary impacts on crop growth. Thus, we aim to classify the seedlings of crop and weed species. In this work, we propose a plant seedlings classification using the benchmark plant seedlings dataset. The dataset contains the images of 12 different species where three belongs to plant species and the other nine belongs to weed species. We implement the classification framework using three different deep convolutional neural network architectures, namely ResNet50V2, MobileNetV2 and EfficientNetB0. We train the models using transfer learning and compare the performance of each model on a test dataset of 833 images. We compare the three models and demonstrate that the EfficientNetB0 performs better with an average F1-Score of 96.26% and an accuracy of 96.52%.

Domestic peat pots biodegradation

Biodegradable nutrient media based on cellulose fibers and peat with added protective materials and stimulating represent the higher form of transplant used in current technologies for producing plant seedlings. Worldwide intensive development of the pots production industry (the effect of increasing demands from agriculture and their quality level in terms of degradability and operational sustainability, with outstanding results in reducing stress factors) led to fast expansion of their use with a high economic efficiency. Determination of potential and effective degradation was achieved by the method for determining cellulolytic activity in soil. Biodegradation potential recorded differences during the course of experience in the sense that it increases as the pots exposure time to treatment is higher. In conclusion, the pots treatment (planted/without plant) has a great influence on the effective biodegradation.

An Innovative Approach to Automated Photo-Activation of Crop Acreage Using UAVs to Stimulate Crop Growth

Photoactivation of plants by laser treatment is a promising direction in the development of modern agricultural production. Treatment of plants with radiation with specified characteristics stimulates the development of plants, the formation of generative traits and an increase in yield. An approach based on the use of a specialized laser installation mounted on an unmanned aerial vehicle (UAV) is proposed to automate the process of photoactivation of large cultivated areas. It is possible to perform laser activation of large areas with minimal expenditure of time and human resources due to autonomous processing of the field with the help of UAVs. An algorithm for calculating a covering trajectory for covering large rectangular areas with a laser spot with given characteristics is proposed in the paper. A methodology for calculating the required power of the laser installation depending on the altitude and flight time of the UAV is presented. The advantage of the developed approach is its versatility, since this approach takes into account the characteristics of a laser installation and can be used with devices of various types. Depending on the laser parameters, the algorithm builds such a trajectory for the UAV so that the irradiation of plant seedlings is uniform throughout the entire processing process. Field experiments were conducted when the UAV moved along the calculated trajectory at a speed of 0.3 m/s and the average processing time for a field 200 m long and 1 m wide was 9 minutes. The results of field experiments show that laser irradiation on most of the studied crops increased the yield and height of the stand (in cereals - in four out of six crops, in legumes - in four out of five studied crops). The proposed algorithm for constructing a path for uniform laser irradiation of a site takes into account the area of the laser spot to ensure the required radiation characteristics when using any laser installation.

Seedling Functional Types and Cotyledons Shape Some Species of Woody Plant

ABSTRACT Woody plant produce small to large seeds which can affect germination process. Cotyledons as an important food source for early growth of seedlings. The position, function, and type of cotyledons when seeds germinate varies. This research aimed to determine cotyledons characters of woody plant seedlings collection in Bogor Botanic Gardens. The position, function and type of cotyledons were observed for 65 species of seedlings.This research used an observative method and was analyzed descriptively.The position of cotyledons when the seeds germinate are epigeal (above the ground) and hypogeal (remain in the soil). The epigeal cotyledons type found in 44 species that is higher than hypogeal cotyledons type (21 species). The function of cotyledons for seedlings is to reserve food or photosynthesis. Species with cotyledon reserve type were more than foliar cotyledons type (43 species and 22 species, respectively). The proportion of seedling types CER (cryptocotylar-epigeal-reservoir), CHR (cryptocotylar-hypogeal-reservoir), PEF (phanerocotylar-epigeal-foliaceous), PER (phanerocotylar-epigeal-reservoir), PHR (phanerocotylar-hypogeal-reservoir) were 3.08%, 16.92%, 33,85%, 29,23%, 16,92%, respectively.

Weed Classification Using Explainable Multi-Resolution Slot Attention

In agriculture, explainable deep neural networks (DNNs) can be used to pinpoint the discriminative part of weeds for an imagery classification task, albeit at a low resolution, to control the weed population. This paper proposes the use of a multi-layer attention procedure based on a transformer combined with a fusion rule to present an interpretation of the DNN decision through a high-resolution attention map. The fusion rule is a weighted average method that is used to combine attention maps from different layers based on saliency. Attention maps with an explanation for why a weed is or is not classified as a certain class help agronomists to shape the high-resolution weed identification keys (WIK) that the model perceives. The model is trained and evaluated on two agricultural datasets that contain plants grown under different conditions: the Plant Seedlings Dataset (PSD) and the Open Plant Phenotyping Dataset (OPPD). The model represents attention maps with highlighted requirements and information about misclassification to enable cross-dataset evaluations. State-of-the-art comparisons represent classification developments after applying attention maps. Average accuracies of 95.42% and 96% are gained for the negative and positive explanations of the PSD test sets, respectively. In OPPD evaluations, accuracies of 97.78% and 97.83% are obtained for negative and positive explanations, respectively. The visual comparison between attention maps also shows high-resolution information.

A novel two-stage method of plant seedlings classification based on deep learning

In agricultural production, weed removal is an important part of crop cultivation, but inevitably, other plants compete with crops for nutrients. Only by identifying and removing weeds can the quality of the harvest be guaranteed. Therefore, the distinction between weeds and crops is particularly important. Recently, deep learning technology has also been applied to the field of botany, and achieved good results. Convolutional neural networks are widely used in deep learning because of their excellent classification effects. The purpose of this article is to find a new method of plant seedling classification. This method includes two stages: image segmentation and image classification. The first stage is to use the improved U-Net to segment the dataset, and the second stage is to use six classification networks to classify the seedlings of the segmented dataset. The dataset used for the experiment contained 12 different types of plants, namely, 3 crops and 9 weeds. The model was evaluated by the multi-class statistical analysis of accuracy, recall, precision, and F1-score. The results show that the two-stage classification method combining the improved U-Net segmentation network and the classification network was more conducive to the classification of plant seedlings, and the classification accuracy reaches 97.7%.

Effects of Cenococcum Geophilum on the Growth of Seedlings Of Ostryopsis Davidiana Dence. And Pinus Tabuleaformis Carr.

The affinity and promoting ability of the Cenococcum geophilum strains (CgSO1, CgSB2, CgO5, SPOP2 and Cg5 #) to Ostryopsis davidiana and Pinus tabulaeformis were studied. The results showed that the tested strains formed ectomycorrhizae（ECM）with O. davidiana except the Cg5 # strain and the infection rates were 40-50%. CgSO1 and CgSPOP2 formed ECM with P. tabulaeformis and the infection rates were 10.3 and 12.4%, respectively. C. geophilum can promote the growth of the two host plant seedlings, especially for root growth. Results proved that the affinity of different strains to host plants and the affinity of the same strain to different host plants are different. Bangladesh J. Bot. 50(3): 679-683, 2021 (September)

PENGARUH PEMBERIAN PUPUK HAYATI MIKORIZA TERHADAP PERTUMBUHAN KAYU PUTIH (Melaleuca leucadendron Linn.)

The purpose of this study was to analyze the effect of presenting mycorrhizae on the growth of eucalyptus plant (Melaleuca leucadendron Linn), to analyze the proper mycorrhizal dose for the growth of eucalyptus plant. This research was conducted in Cindai Alus Village, Martapura District, Banjar Regency, Kalimantan selatan. The implementation of this study took 4 months. The experimental design used was CRD (Completely Randomized Design) by giving mycorrhizae whit 5 treatments, each treatment with 20 replications, so that all of them had 100 experimental units.The results showed that the setetage of live eucalyptus plant seedlings in each treatment mycorrhizae was 100%. , and offered eucalyptus plant mycorrhizae had a very significant effect on increasing height, increasing the number of shoots, but not affecting the increase in diameter. The administration of 40 grams of mycorrhizae had the best effect on increasing height (9.55 cm), number of ban (11.75) and increasing diameter (0.493 mm). Tujuan dari penelitian ini adalah untuk menganalisis pengaruh pemberian mikoriza pada pertumbuhan kayu putih (Melaleuca leucadendron Linn), menganalisis dosis mikoriza yang tepat untuk pertumbuhan kayu putih. Penelitian ini dilakukan di Desa Cindai Alus, Kecamatan Martapura, Kabupaten Banjar, Kalimantan Selatan. Implementasi penelitian ini membutuhkan 4 bulan. Desain eksperimental yang digunakan adalah RAL (Rancangan Acak Lengkap) dengan pemberian mikoriza dan 5 perlakuan, masing-masing perlakuan dengan 20 ulangan, sehingga semuanya memiliki 100 unit percobaan, Hasil pengamatan menunjukkan bahwa persentase semai kayu putih hidup dalam setiap perlakuan pemberian mikoriza adalah 100%, dan pemberian mikoriza bibit kayu putih memiliki efek yang sangat signifikan terhadap peningkatan tinggi, peningkatan jumlah tunas, tetapi tidak mempengaruhi peningkatan diameter. Pemberian mikoriza sebanyak 40 gram memberikan efek terbaik pada peningkatan tinggi (9,55 cm), jumlah tunas (11, 75 bua) dan peningkatan diameter (0,493 mm).

An illuminated respiratory activity monitoring system identifies priming-active compounds in plant seedlings

Background Growing large crop monocultures and heavily using pesticides enhances the evolution of pesticide-insensitive pests and pathogens. To reduce pesticide use in crop cultivation, the application of priming-active compounds (PrimACs) is a welcome alternative. PrimACs strengthen the plant immune system and could thus help to protect plants with lower amounts of pesticides. PrimACs can be identified, for example, by their capacity to enhance the respiratory activity of parsley cells in culture as determined by the oxygen transfer rate (OTR) using the respiration activity monitoring system (RAMOS) or its miniaturized version, µRAMOS. The latter was designed for with suspensions of bacteria and yeast cells in microtiter plates (MTPs). So far, RAMOS or µRAMOS have not been applied to adult plants or seedlings, which would overcome the limitation of (µ)RAMOS to plant suspension cell cultures. Results In this work, we introduce a modified µRAMOS for analysis of plant seedlings. The novel device allows illuminating the seedlings and records the respiratory activity in each well of a 48-well MTP. To validate the suitability of the setup for identifying novel PrimAC in Arabidopsis thaliana, seedlings were grown in MTP for seven days and treated with the known PrimAC salicylic acid (SA; positive control) and the PrimAC candidate methyl 1-(3,4-dihydroxyphenyl)-2-oxocyclopentane-1-carboxylate (Tyr020). Twenty-eight h after treatment, the seedlings were elicited with flg22, a 22-amino acid peptide of bacterial flagellin. Upon elicitation, the respiratory activity was monitored. The evaluation of the OTR course reveals Tyr020 as a likely PrimAC. The priming-inducing activity of Tyr020 was confirmed using molecular biological analyses in A. thaliana seedlings. Conclusion We disclose the suitability of µRAMOS for identifying PrimACs in plant seedlings. The difference in OTR during a night period between primed and unprimed plants was distinguishable after elicitation with flg22. Thus, it has been shown that the µRAMOS device can be used for a reliable screening for PrimACs in plant seedlings.