Classification Of Plants Research Articles

Exploring the interspecific relationships among Dendrobium species via electrochemical fingerprinting techniques

This study investigated the interspecific relationships among 29 Dendrobium species using electrochemical fingerprinting techniques. Differential pulse voltammetry was employed to record species-specific electrochemical profiles in both phosphate buffer solution (PBS) and acetic acid buffer solution (ABS). Distinct voltammetric patterns were observed for each species, with ABS yielding higher peak currents compared to PBS. On average, peak currents in ABS were 37.2 % ± 5.8 % higher than in PBS across all species. For example, D. nobile showed a peak current of 12.3 μA in ABS compared to 8.9 μA in PBS, a 38.2 % increase. Normalized two-dimensional plots of the electrochemical fingerprints served as unique signatures for species differentiation. Cluster analysis of the electrochemical data revealed phylogenetic relationships largely consistent with previous molecular studies, while also highlighting novel groupings. For instance, D. parishii and D. pulchellum formed separate branches, indicating distant relationships with other species. Common species such as D. officinale, D. nobile, D. primulinum, and D. loddigesii clustered together, confirming their close relationships. Interestingly, species from Sect. Dendrobium showed mixed clustering with species from four other sections, reflecting the genus's complex evolutionary history. While some inconsistencies with previous molecular phylogenetic studies were noted, such as the distant relationship between D. parishii and D. anosmum, the overall congruence between electrochemical and molecular data validates the potential of this approach. This study demonstrates that electrochemical fingerprinting, combined with appropriate data analysis techniques, can provide valuable insights into the taxonomic relationships and chemical diversity of Dendrobium species, offering a complementary tool to traditional morphological and molecular methods for plant systematics and conservation efforts.

A Multitask Learning-Based Vision Transformer for Plant Disease Localization and Classification

Plant disease detection is a critical task in agriculture, essential for ensuring crop health and productivity. Traditional methods in this context are often labor-intensive and prone to errors, highlighting the need for automated solutions. While computer vision-based solutions have been successfully deployed in recent years for plant disease identification and localization tasks, these often operate independently, leading to suboptimal performance. It is essential to develop an integrated solution combining these two tasks for improved efficiency and accuracy. This research proposes the innovative Plant Disease Localization and Classification model based on Vision Transformer (PDLC-ViT), which integrates co-scale, co-attention, and cross-attention mechanisms and a ViT, within a Multi-Task Learning (MTL) framework. The model was trained and evaluated on the Plant Village dataset. Key hyperparameters, including learning rate, batch size, dropout ratio, and regularization factor, were optimized through a thorough grid search. Early stopping based on validation loss was employed to prevent overfitting. The PDLC-ViT model demonstrated significant improvements in plant disease localization and classification tasks. The integration of co-scale, co-attention, and cross-attention mechanisms allowed the model to capture multi-scale dependencies and enhance feature learning, leading to superior performance compared to existing models. The PDLC-ViT model evaluated on two public datasets achieved an accuracy of 99.97%, a Mean Average Precision (MAP) of 99.18%, and a Mean Average Recall (MAR) of 99.11%. These results underscore the model's exceptional precision and recall, highlighting its robustness and reliability in detecting and classifying plant diseases. The PDLC-ViT model sets a new benchmark in plant disease detection, offering a reliable and advanced tool for agricultural applications. Its ability to integrate localization and classification tasks within an MTL framework promotes timely and accurate disease management, contributing to sustainable agriculture and food security.

Genetic and morphological evidence support the specific status of the endemic Ericaandevalensis (Ericales, Ericaceae).

Assessing the taxonomic status of closely related taxa is crucial in plant systematics and can have important implications for conservation and human plant use. Ericaandevalensis Cabezudo & Rivera is a metallophyte endemic species from highly metal-polluted soils of SW Iberian Peninsula, an area with a mining history going back more than 5,000 years. Ericaandevalensis is closely related to Ericamackayana Bab., a northern Iberian species also present in western Ireland. The status of E.andevalensis as a species or subspecies subordinated to E.mackayana is subject to debate. Here, we assessed the genetic and phenotypic relationship between both species, including the population structure of E.andevalensis. We used high throughput sequencing to determine genome-wide Single Nucleotide Polymorphisms (SNPs), and morphometric analyses from 35 reproductive and vegetative traits. The morphological analysis showed at least eight characters that can discriminate the two species, from which ovary hairiness and the size of leaf glandular hairs were the most informative. Genetic analyses showed that each species formed a monophyletic cluster with full support, separated by an interspecific genetic distance >4-fold higher than intra-specific distance. Population genetic analyses of E.andevalensis shows that populations are highly structured, with the Portuguese one as the most isolated and less variable. These results support the recognition of E.andevalensis as a distinct species with a highly constrained ecological requirements and a narrow geographic distribution, but with a limited gene flow between populations. We discuss the implications of these outcomes in conservation policies and potential uses of E.andevalensis such as decontamination of polluted soils.

Masked autoencoder-based self-supervised learning for forest plant classification

Masked autoencoder-based self-supervised learning for forest plant classification

A novel and optimized IoT –ML based plant classification, monitoring and prediction system

A novel and optimized IoT –ML based plant classification, monitoring and prediction system

Complete chloroplast genomes and phylogenetic relationships of Pedicularis chinensis and Pedicularis kansuensis

The complete cp genomes of Pedicularis chinensis (GenBank accession number: OQ587614) and Pedicularis kansuensis (GenBank accession number: OQ587613) were sequenced, assembled, and annotated. Their chloroplast (cp) genome lengths were 146,452 bp, and 146,852 bp, respectively; 120 and 116 genes were identified, comprising 75 and 72 protein-coding genes (PCGs), 37 and 36 transfer RNA (tRNA) genes, and 8 and 8 ribosomal RNA (rRNA) genes, for P. chinensis and P. kansuensis, respectively. A simple sequence repeat (SSR) analysis revealed that the repetitive sequences were mainly composed of mononucleotide repeats (A/T motif) and dinucleotide repeats (AT/TA motif). Comparative genomics identified several variant genes (rpl22, rps19, rpl12, ycf1, trnH, psbA, and ndhH) and variant regions (trnS-GGA, trnV-UAC, ndhJ-trnV, ycf4-cemA, ndhE-nhdG, and rpl32-trnL) with a high Pi, indicating the potential to serve as deoxyribo nucleic acid (DNA) barcodes for Pedicularis species identification. The results show that the cp genomes of P. chinensis and P. kansuensis were the same as those of other plants in Pedicularis, with different degrees of AT preference for codons. Large differences in the number of SSRs and the expansion of the inverted repeat (IR) region showed strong variability and interspecific differentiation between these two species and other species represented in the genus Pedicularis. A phylogenetic analysis showed that P. kansuensis had the closest relationship with P. oliveriana, and P. chinensis had the closest relationship with P. aschistorhyncha. These results will facilitate the study of the phylogenetic classification and interspecific evolution of Pedicularis plants.

Leaf Segmentation Using Modified YOLOv8-Seg Models.

Computer-vision-based plant leaf segmentation technology is of great significance for plant classification, monitoring of plant growth, precision agriculture, and other scientific research. In this paper, the YOLOv8-seg model was used for the automated segmentation of individual leaves in images. In order to improve the segmentation performance, we further introduced a Ghost module and a Bidirectional Feature Pyramid Network (BiFPN) module into the standard Yolov8 model and proposed two modified versions. The Ghost module can generate several intrinsic feature maps with cheap transformation operations, and the BiFPN module can fuse multi-scale features to improve the segmentation performance of small leaves. The experiment results show that Yolov8 performs well in the leaf segmentation task, and the Ghost module and BiFPN module can further improve the performance. Our proposed approach achieves a 86.4% leaf segmentation score (best Dice) over all five test datasets of the Computer Vision Problems in Plant Phenotyping (CVPPP) Leaf Segmentation Challenge, outperforming other reported approaches.

Cryptovivipary mechanism in the wild caper bush Capparis sepiaria (Capparaceae)

Cryptovivipary is the process in which seeds germinate inside the fruit but the seedling or plumule comes out when the fruit detaches from the mother plant. This paper highlights the details of the mechanism of cryptovivipary in the wild caper bush. In addition, an updated classification of viviparous plants is provided to better describe and understand this trait.

Deep learning-based restoration of nonlinear motion blurred images for plant classification using multi-spectral images

There have been various plant image-based studies for segmentation, deblurring, super-resolution reconstruction, and classification. However, nonlinear motion blur in thermal images was not considered in the existing studies on plant classification. Nonlinear motion blur occurs in images due to camera or plant movements, and it causes the degradation of plant classification accuracy. Moreover, nonlinear motion blur in images gets worse when both camera and plant movements occur simultaneously. In this case, it becomes difficult to recognize plants, and the performance of plant image classification becomes very low. Therefore, to reduce the nonlinear motion blur, a thermal and visible light plant images-based deblurring network (TVPD-Net) is proposed in this study. In addition, a thermal and visible light plant images-based classification network (TVPC-Net) is also proposed to improve the plant classification performance on deblurred images. Experimental results revealed that the proposed TVPD-Net achieved 21.21 and 22.53 of the peak signal-to-noise ratio (PSNR), and 0.726 and 0.737 of the structural similarity index measure (SSIM) on both visible light and thermal plant image datasets which were self-collected, respectively. Moreover, the proposed TVPC-Net with deblurred images by TVPD-Net achieved 92.52 % (top-1 accuracy) and 87.73 % (harmonic mean of precision and recall (F1-score)). In addition, the experimental results on an open dataset named Hyperspectral Flower Dataset (HFD100) revealed that the proposed plant classification method achieved 90.94 % of top-1 accuracy and 86.21 % of F1-score. The accuracies of the proposed methods are greater than those of the state-of-the-art methods.

An Improved Ningxia Desert Herbaceous Plant Classification Algorithm Based on YOLOv8.

Wild desert grasslands are characterized by diverse habitats, uneven plant distribution, similarities among plant class, and the presence of plant shadows. However, the existing models for detecting plant species in desert grasslands exhibit low precision, require a large number of parameters, and incur high computational cost, rendering them unsuitable for deployment in plant recognition scenarios within these environments. To address these challenges, this paper proposes a lightweight and fast plant species detection system, termed YOLOv8s-KDT, tailored for complex desert grassland environments. Firstly, the model introduces a dynamic convolutional KernelWarehouse method to reduce the dimensionality of convolutional kernels and increase their number, thus achieving a better balance between parameter efficiency and representation ability. Secondly, the model incorporates triplet attention into its feature extraction network, effectively capturing the relationship between channel and spatial position and enhancing the model's feature extraction capabilities. Finally, the introduction of a dynamic detection head tackles the issue related to target detection head and attention non-uniformity, thus improving the representation of the target detection head while reducing computational cost. The experimental results demonstrate that the upgraded YOLOv8s-KDT model can rapidly and effectively identify desert grassland plants. Compared to the original model, FLOPs decreased by 50.8%, accuracy improved by 4.5%, and mAP increased by 5.6%. Currently, the YOLOv8s-KDT model is deployed in the mobile plant identification APP of Ningxia desert grassland and the fixed-point ecological information observation platform. It facilitates the investigation of desert grassland vegetation distribution across the entire Ningxia region as well as long-term observation and tracking of plant ecological information in specific areas, such as Dashuikeng, Huangji Field, and Hongsibu in Ningxia.

AI-Enhanced Medicinal Plant Identification System with Multilingual Social Media Integration

Sri Lanka is a country with a Ayurvedic culture which cannot be experienced anywhere in the world. This cultural system is based on a series of knowledge passed on from generations over 3000 years that could treat a variety of diseases. This traditional ayurvedic system consist of a vast herbal plant collection. Most of the information about this ayurvedic system is written in manuscripts for thousands of years. Sri Lanka lacks a proper system which is specific to ayurvedic sector is a major concern that should be addressed at present. Absence of a system has lead to problems and difficulties in identification and classification of herbal plants, to transfer knowledge about herbal plants and to conserve these ayurvedic plants for the future generation. Another concern is that Ayurvedic undergraduate students face many difficulties when gathering knowledge of these herbal plants and medicinal practices. Sri Lanka does not comprise with a full ayurvedic plant inventory system is another major concern that identified in the country. By considering all the problems an intelligent system has been recognized as a solution. The system will be based on Deep Learning, CNN, GIS, Artificial Intelligence and Machine Learning based principals to cater all the identified problems. The system will be able to identify ayurvedic plant with an image of a leave, flower, or fruit as input. And also, system will be able to classify and provide a detailed description about the identified plant including medicinal value and the distribution of the plant in the island. System will provide a crowdsourcing social media facility with both English and Sinhala languages to share information with fellow herbalist in the country.

Medicinal and poisonous plants classification from visual characteristics of leaves using computer vision and deep neural networks

Poisonous plants are the third largest category of poisons known globally, which pose a risk of poisoning and death to humans. Currently, the identification of medicinal and poisonous plants is done by humans using experimental methods, which are not accurate and are associated with many errors, and also the use of laboratory methods requires experts and this method is very costly and time-consuming. Therefore, distinguishing between medicinal and poisonous plants is very important using emerging, non-destructive, fast and accurate methods such as computer vision and artificial intelligence. In this study, we propose a robust and generalized model using spatial attention (SA) and channel attention (CA) modules for the classification of different plants. A dataset containing 900 confirmed images of three plant classes (oregano, poisonous and weed) was used. The attention mechanisms enhance efficiency of deep learning (DL) networks by allowing them to precisely focus on all relevant input elements. In order to enhance the performance of the proposed model, the CA was implemented based on four pooling operations including global average pooling-based CA (GAP-CA), mixed pooling-based CA (Mixed-CA), gated pooling-based CA (Gated-CA), and tree pooling-based CA (Tree-CA) operations. The results showed that the DL model based on Tree-CA had promising performance and outperformed other state-of-the-art models, achieving the values of 99.63%, 99.38%, 99.52%, 99.74%, and 99.42%, for accuracy, precision, recall, specificity, and F1-score, respectively. The findings support our proposed attention model's success in identifying medicinal plants from similar poisonous plants. Recent advancements in computer-based technologies and artificial intelligence enable automatic detection of medicinal and poisonous plants, revolutionizing traditional identification methods.

Exploring the micromorphological diversity of palynomorphic flora from lesser Himalaya biodiversity hotspot

Palynology, a prominent field in plant systematics and biodiversity studies, plays a vital role in identifying and determining the plant species present in a specific region. The current study was performed to evaluate the micromorphological traits of pollen from flora of Lesser Himalaya. Pollen microstructural variations aid in the identification of species belonging to specific botanical families and various geographic habitats. Flowers of 24 selected species categorized into 16 families were collected, preserved and then acetolysis protocol followed. Pollen was examined under a light and scanning microscopy (LM and SEM) for palynomorph description. The palynomorphs characteristics such as size, shape, exine surface, and aperture orientation, were examined. Status of these plants show that herbs are being dominant (11 species), while shrubs (7 species), climbers (3 species), bulbous plants (2 species), small tree, sedge, weed (1 species each). Pollen shape determined in equatorial view were; spheroidal, sub-prolate, oblate-spheroidal, prolate, spherical and sub-oblate. The variations were seen among pollen types; tricolpate, tricolporate and polyporate in most of the species. Exine stratification was observed mostly scabrate while echinate, cristate-reticulate, granulate, punctate, rugulate-perforate, striate-rugulate, verrucate, cristate-foveolate was visualized in each different species. Palynomorph apertural patterns were observed sunken, furrowed, slightly bulged, scabrate, granulate, slit like, and perforate. The largest polar diameter was measured in Hymenocallis littoralis (138.6 µm) whereas smallest in Parthenium hysterophorus (14.70 µm). Equatorial distance was calculated maximum for Cascabela thevetia (110.1 µm) and minimum for Hibiscus rosa-sinensis (1.7 µm). P/E ratio was calculated largest in Hymenocallis littoralis (1.8) and lowest in Duranta erecta (0.89). The palynomorphs taxonomic characters investigated can be helpful in species level identification and provide a baseline to conduct more systematic research with respect to specific plant families and genera.

Characterization of the plastid and mitochondrial genomes of Aeschynomene indica (Fabaceae)

The current study undertook a detailed analysis of both the plastid and mitochondrial genomes (plastome and mitogenome) of Aeschynomene indica (Fabaceae), a plant notable both as a problematic weed in global rice cultivation and for its medicinal properties. The primary goal was to furnish comprehensive organelle genome resources and explore genomic characteristics. A sample of A. indica was collected and sequenced using a next-generation sequencing technique. The resulting plastome is 160,502 base pairs (bp) long and includes 128 genes—comprising 83 protein-coding genes, 37 transfer RNA (tRNA) genes, and eight ribosomal RNA (rRNA) genes. The plastome displays a typical quadripartite structure consisting of a large single-copy region (89,347 bp), a small single-copy region (19,803 bp), and two inverted repeat regions (each 25,676 bp). The mitogenome spans 388,249 bp and contains 57 genes, which include 34 protein-coding genes, 20 tRNA genes, and four rRNA genes. Phylogenetic analyses leveraging these organelle genome sequences produced well-resolved trees, elucidating the phylogenetic position of A. indica within the Fabaceae. The organelle genomes of A. indica presented in this study offer valuable genomic resources that will facilitate further research in plant systematics and biotechnology.

MTJNet: Multi-task joint learning network for advancing medicinal plant and leaf classification

Identifying and classifying medicinal plants is crucial for pharmaceutical research, traditional medicine, and biodiversity conservation. However, manual classification is time-consuming, error-prone, and requires expert knowledge. This problem statement highlights the need for an automated, efficient, and accurate method by which to distinguish between different species and ensure the quality and safety of medicinal plant leaves. The state-of-the-art methods employ only leaf or plant images for the classification; however, these methods work in particular scenarios (only leaf or plant). In this study, we first integrate the dataset of medicinal plants with corresponding leaves to develop a generalizable model and then apply pre-processing techniques (using vein morphometric features and Sobel edge detection) to highlight the significant characteristics in the images. Then, we propose a novel multi-task joint learning network (MTJNet) for classifying medicinal plant leaves. The proposed MTJNet includes local and global feature extractors that retrieve prominent features for robust classification. Further, we blend the features extracted from the local and global networks to enhance the contextual features. Last, we assign the embedded features to dense layers for additional feature extraction and classification. To evaluate the proposed MTJNet, we employ an Indian medicinal leaf dataset and achieve a precision of 99.60%, a recall of 99.62%, an accuracy of 99.71%, and an F1 score of 99.58%. The experimental results show that the MTJNet statistically defeats prevalent models, thereby proving its effectiveness for medical and industrial applications.

An intelligent deep residual learning framework for tomato plant leaf disease classification

Modern agriculture has been fascinated by various advancements in agriculture and the processing of foods and supply management that provision farmers to improve production. The health of plants is essential to improve production and economic growth. Diseases in plants can affect production and create a rigorous impact on the quality and create a hazard to food safety. Hence, detecting and classifying plant leaf diseases is essential to prevent the disease spread across the plants in the agriculture field and to improve productivity. The researchers in existing frameworks utilized artificial intelligence and machine learning techniques to demonstrate noteworthy solutions. However, a few issues exist related to noises in the images, hyperparameter selection problems, and over-fitting problems that influence prediction accuracy. The proposed model jellyfish ResNet(JF-ResNet) works well to achieve a better accuracy level by incorporating jellyfish optimized ResNET for tomato plant leaf disease identification and classification. The performance metrics such as Accuracy, specificity, sensitivity, and F1-Score is used to evaluate the performance of the JF-ResNet model. The proposed model achieves 97.3% accuracy, 95.3% sensitivity, 96.1% specificity, 96.9% recall, 96.4% precision and 97.1% F1-Score.

Identification of Different Medicinal Plants / Raw materials through PCA Using Support Vector Machine and K-Means Clustering Algorithms

This research explores the identification of plants by analyzing their leaves, utilizing digital image processing techniques to examine shape, color, and texture features. Machine learning algorithms are then applied to classify the plants based on these characteristics. The study involves an examination of various image processing methods integrated with machine learning to differentiate between plant species using leaf features captured in images. Image processing serves as the primary approach for categorizing plants based on unique leaf characteristics or specific leaf regions, enabling subsequent identification through image analysis. The research specifically focuses on the identification and classification of plant species based on the analysis of different leaf parts. The paper presents an overview of classification methods applied in plant leaf identification, with Support Vector Machine (SVM) being the primary algorithm used for plant identification and classification. Experimental results demonstrate enhanced accuracy in plant species identification based on leaf features, offering a more efficient and time-saving approach to plant identification

Plant Leaf Disease Detection and Suggesting Medication using Image Processing and Convolutional Neural Network Techniques

In day-to-day life many factors that affect plants. Many problems are occurring at a rapid pace and new diseases are rapidly being identified. In today’s world of pollution and soil quality, disease detection is essential and its supplement is required to protect the plants. The main motivation of doing this project is to present a disease prediction model for the prediction of occurrence of plant crop disease. Further, this research work is aimed towards identifying the best classification algorithm for identifying the possibility of disease in plants. The identification of the possibility of disease in plants is complicated task for practitioners because it requires years of experience and intense tests to be conducted. The main objective of this significant research work is to identify the best classification algorithms suitable for providing maximum accuracy when classification of normal and abnormal plants is carried out. Convolutional neural network (CNN) architecture is used to map the relationship between the indoor PM and weather data to the found values. The proposed method is compared with the state-of-the-art deep neural network (DNN) based techniques in terms of the root mean square and mean absolute error accuracy measures. The applied CNN classification helps to predict the heart disease with more accuracy in the new data set. The coding language used is Python 3.10..

Introduction of Fruit and Nut Crops in Medieval India

India grows a large variety of fruits and nuts (hard-shelled fruit or seed) which play an important role in the economy of the country. A large number of fruit and nut yielding species grown in India are introductions during the medieval period (1200-1800 A.D.) from foreign countries and centres of origin of cultivated plants. Some of these species introduced have been successful and later became important export crops such as cashewnut and groundnut. Much of the information on the early history of plant introduction is scattered in archaeological accounts, ethnographies, gazetteers and travelogues of several visitors and dignitaries to India, and in the classical works, memoirs and autobiographies of Moghul emperors published during the medieval period. The Present account deals with the introduction of fruit and nut crops in India during the medieval period. (1200-1800 A.D.). There is need for collaborative work between the plant systematist, horticulturist, linguist and biohistorian for compilation of an authenticated history of pomology and horticulture in India.

Classification and Phylogeny

A scientific study of classification of plants started with Linnaeus when in 1753 he published his monumental 'Species Plantarum' that made him the father of systematic botany. Linnaeus' system of classification was an artificial one, based as it was on tome arbitrarily selected characters.' He himself regarded it as "one of convenience until the tíme when a. Natural System could take-its place". Subsequently, however, classification began to be based on general resemblances and differences and the species came to be regarded as a concept rather than a fixed entity. Such a classification came to be known as a 'natural', 'logical' or more recently 'general' classification.It is emphasized that in the present state of our knowledge it is almost impossible to trace any group phylogenies. Even if we are able to determine them it will be very difficult to incorporate them in our classificatory schemes. It is therefore concluded that the main task of the systematist should be "to make a general classification which shall express as far as possible in rational order all that is known concerning plants and animals. This ideal which, even if never attained, is one which may still make the systematist proud in the magnitude of his task. It is an ideal greater than the phylogenetic ideal which is included in it and one which in the process of attempted attainment must make taxonomy what it should be, the focal point of biology".