Classification Of Plants Research Articles

Nuclear ITS and AFLPs provide surprising implications for the taxonomy of Tephroseris longifolia agg. and the endemic status of T. longifolia subsp. moravica

The proportion of endemic taxa is a frequently used measure for assesment of biodiversity hotspots within the Carpathians. For confirming of an endemic status of a taxon, a thorough evaluation using modern tools of plant systematics is desirable. Tephroseris longifolia subsp. moravica has been considered to be a Western Carpathian endemic, but a previous study of morphological and genome size differentiation within the T. longifolia agg. showed that its endemic position is controversial. Therefore, in this study we aimed to detect genetic variability and evolutionary relationships within the T. longifolia agg. using nuclear ITS sequences and highly variable AFLPs on 38 populations covering the distribution range of the aggregate. Additionally, we analysed 22 populations of other putatively related species (T. aurantiaca, T. capitata, T. crispa, T. integrifolia and T. papposa). Genetic analyses proved that T. longifolia agg. is genetically well differentiated, but the results contradict the current taxonomic concept. Tephroseris longifolia aggregate, as defined here, comprises T. longifolia, T. crispa, T. pseudocrispa, T. tenuifolia and T. italica. Within T. longifolia, two subspecies can be recognised, namely T. l. subsp. longifolia and T. l. subsp. moravica. Furthermore, AFLPs confirm the close relationship of Pannonian and one Croatian populations to T. l. subsp. moravica. Therefore, we assume that T. l. subsp. moravica should not be considered a Western Carpathian endemic and most probably migrated to the Western Carpathian area during the post-glacial period.

Real-time classification of weeds in organic carrot production using deep learning algorithms

This paper proposes a real-time machine learning approach for carrot plants classification in organic farming using Convolutional Neural Network. Artificial neural networks become increasingly popular for image processing tasks, e.g. the classification of complex structures in images. The problem is not very often the accuracy of the classification, but the speed of calculation. The core of this paper presents a real-time calculation flow for the neural networks, in which all the individual steps are summarized. It also briefly discusses the used sensors, which are suitable for the Convolutional Neural Network and the pre-processing which extracts the plants from the background in order to keep the load on the neural network as low as possible.

Botany at Stefan Batory University in Vilna (Wilno, Vilnius) (1919–1939)

The university in Vilna (in Polish: Wilno, now: Vilnius, Lithuania), founded in 1579, by Stefan Batory (Stephen Báthory), King of Poland and Grand Duke of Lithuania, was a centre of Polish botany in 1780–1832 and 1919–1939. In the latter period the university functioned under the Polish name Uniwersytet Stefana Batorego (in English: Stefan Batory University). It comprised six departments connected with botany (General Botany, Pharmacognosy and Cultivation of Medicinal Plants, Plant Taxonomy, Botanical Garden, Garden of Medicinal Plants, and Natural History Museum). There worked such distinguished scientists, as: Jakub Mowszowicz (1901–1983), phytogeographer and phytosociologist; Jan Muszyński (1884–1957), botanist and pharmacist; Bronisław Szakien (1890–1938), cytologist and mycologist; Piotr Wiśniewski (1881––1971), physiologist; and Józef Trzebiński (1867–1941), mycologist and phytopathologist. Ca. 300 publications (including ca. 100 scientific ones) were printed in the period investigated, dealing mainly with morphology and anatomy, cytology, plant physiology, floristics (floristic geography of plants), systematics (taxonomy) of vascular plants, mycology and phytopathology, ecology of plant communities (phytosociology), as well as ethnobotany, and history of botany. Stefan Batory University was also an important centre of teaching and popularization of botany in that region of Europe. The aim of the article is to describe the history of botany at the Stefan Batory University in 1919–1939. Botanika na Uniwersytecie Stefana Batorego w Wilnie (Vilna, Vilnius) (1919–1939) Abstrakt Uniwersytet w Wilnie (w języku angielskim: Vilna, obecnie: Vilnius w Republice Litewskiej), założony w 1579 r. przez Stefana Batorego, króla Polski i wielkiego księcia Litwy, był ośrodkiem polskiej botaniki w latach 1780–1832 oraz 1919–1939. W tym ostatnim okresie funkcjonował pod nazwą Uniwersytet Stefana Batorego (w języku angielskim: Stefan Batory University). W latach 1919–1939 zorganizowano następujące zakłady związane z botaniką: Botaniki Ogólnej, Farmakognozji i Hodowli Roślin Lekarskich, Systematyki Roślin, Ogród Botaniczny, Ogród Roślin Lekarskich oraz Muzeum Przyrodnicze. W ośrodku wileńskim pracowali wybitni uczeni, m.in. Jakub Mowszowicz (1901–1983), fitogeograf i fitosocjolog; Jan Muszyński (1884–1957), botanik i farmaceuta; Bronisław Szakien (1890–1938), cytolog i mykolog; Piotr Wiśniewski (1881–1971), fizjolog oraz Józef Trzebiński (1867–1941), mykolog i fitopatolog. Badacze roślin ogłosili drukiem ok. 300 publikacji (w tym ok. 100 naukowych) dotyczących głównie morfologii i anatomii, cytologii, fizjologii roślin, florystyki (florystycznej geografii roślin), systematyki (taksonomii) roślin naczyniowych, mykologii i fitopatologii, ekologii zbiorowisk roślinnych (fitosocjologii), a także etnobotaniki i historii botaniki. Uniwersytet Stefana Batorego był również ważnym ośrodkiem nauczania i popularyzacji botaniki w tym regionie Europy. Celem artykułu jest opracowanie historii botaniki na Uniwersytecie Stefana Batorego w latach 1919–1939.

An optimized model based on convolutional neural networks and orthogonal learning particle swarm optimization algorithm for plant diseases diagnosis

Abstract The plant disease classification based on using digital images is very challenging. In the last decade, machine learning techniques and plant images classification tools such as deep learning can be used for recognizing, detecting and diagnosing plant diseases. Currently, deep learning technology has been used for plant disease detection and classification. In this paper, an ensemble model of two pre-trained convolutional neural networks (CNNs) namely VGG16 and VGG19 have been developed for the task plant disease diagnosis by classifying the leaves images of healthy and unhealthy. In this context, CNNs are used due to its capability of overcoming the technical problems which are associated with the classification problem of plant diseases. However, CNNs suffer from a great variety of hyperparameters with specific architectures which is considered as a challenge to identify manually the optimal hyperparameters. Therefore, orthogonal learning particle swarm optimization (OLPSO) algorithm is utilized in this paper to optimize a number of these hyperparameters by finding optimal values for these hyperparameters rather than using traditional methods such as the manual trial and error method. In this paper, to prevent CNNs from falling into the local minimum and to train efficiently, an exponentially decaying learning rate (EDLR) schema is used. In this paper, the problem of the imbalanced used dataset has been solved by using random minority oversampling and random majority undersampling methods, and some restrictions in terms of both the number and diversity of samples have been overcome. The obtained results of this work show that the accuracy of the proposed model is very competitive. The experimental results are compared with the performance of other pre-trained CNN models namely InceptionV3 and Xception, whose hyperparameters were selected using a non-evolutionary method. The comparison results demonstrated that the proposed diagnostic approach has achieved higher performance than the other models.

Classification of Plant leaf diseases: A Deep Learning Method

For the continuous existence of human life, agriculture plays a vital role due to the dependency of lives of people on it for production of food. In order to meet high production rate, precision farming is required. Presently, substantial developments attained “in the field of image processing and recognition” which has being a foremost challenge earlier in the practice of Precision farming. The reduction in the growth, quality and quantity of plants is due to expansion of plant diseases.The survey literature confers various plant leaf diseases and their detections during different phases through various state of art machine and deep learning techniques. The particular emphasis in this paper is on detection and classification of plant “leaf diseases” through deep “convolutional neural networks algorithms.

Use of Support Vector Machine to Classify Rhizomes Based on Color

Rhizome plants are one type of herbal medicinal plants which have various types and benefits. To find out the types of rhizome plants can be done by the classification process. Classification of rhizome plants can be done by identifying the color of the flesh of the rhizome. Identification technology in this study is needed to accelerate and facilitate the process of identifying rhizome species in the form of digital image data. Digital image processing can be used to identify rhizomes based on color using the support vector machine (SVM). In this study, the type of rhizome used as an input image was temulawak, temu hitam and temu mangga. Image preprocessing is done to get the color image of the rhizome flesh which is then used in the feature extraction stage. Image feature extraction process is done to get the color characteristics of the image by calculating the image RGB value. The image color value obtained has been used to identify rhizome species using SVM. The results obtained indicate that the identification of rhizome types using SVM provides an accuracy of 87.5%.

Plant Disease Detection and Classification by Deep Learning.

Plant diseases affect the growth of their respective species, therefore their early identification is very important. Many Machine Learning (ML) models have been employed for the detection and classification of plant diseases but, after the advancements in a subset of ML, that is, Deep Learning (DL), this area of research appears to have great potential in terms of increased accuracy. Many developed/modified DL architectures are implemented along with several visualization techniques to detect and classify the symptoms of plant diseases. Moreover, several performance metrics are used for the evaluation of these architectures/techniques. This review provides a comprehensive explanation of DL models used to visualize various plant diseases. In addition, some research gaps are identified from which to obtain greater transparency for detecting diseases in plants, even before their symptoms appear clearly.

Study on the Deep Learning with Image Recognition to Plant Classification

Because of long cultivation period, the three main issues of the medicinal mushroom Cordyceps militaris need to be solved urgently including high manpower, unstable quantity and quality, high electric power consumption. This article proposes the intelligent technologies of (Artificial intelligence), and Big data analytics in the application to the cultivation of Cordyceps militaris, AI which cost-effectively solve the problems of labor storage, issues of quantity and quality, and energy consumption in the agriculture filed. In this study, the intelligent system for environmental monitoring and management is developed by optimizing light, temperature and humidity of the culture environment of C. militaris regarding of the bio-active ingredients. The new strains with the high bioactive ingredients will be developed with the enhancement of the elicitors and environmental stress in the augmented intelligent supervision system. Furthermore, the combined bio-active ingredients will be developed for new medicines.

Review on Techniques for Plant Leaf Classification and Recognition

Plant systematics can be classified and recognized based on their reproductive system (flowers) and leaf morphology. Neural networks is one of the most popular machine learning algorithms for plant leaf classification. The commonly used neutral networks are artificial neural network (ANN), probabilistic neural network (PNN), convolutional neural network (CNN), k-nearest neighbor (KNN) and support vector machine (SVM), even some studies used combined techniques for accuracy improvement. The utilization of several varying preprocessing techniques, and characteristic parameters in feature extraction appeared to improve the performance of plant leaf classification. The findings of previous studies are critically compared in terms of their accuracy based on the applied neural network techniques. This paper aims to review and analyze the implementation and performance of various methodologies on plant classification. Each technique has its advantages and limitations in leaf pattern recognition. The quality of leaf images plays an important role, and therefore, a reliable source of leaf database must be used to establish the machine learning algorithm prior to leaf recognition and validation.

Tom Beeckman

Tom Beeckman

Hybrid Group Search Optimizer for Plant Leaf Classification

The analysis of Big Data is a data mining discipline in which large quantity of unstructured data is analysed which can be challenging to store and also to retrieve efficiently. The classification of plants is based on the identification of leaf that has a very broad application in both agriculture and medicine. In this work, a method which is computerized is used to recognize a plant leaf based on images is proposed. The proposed method extracts features from the image and these are used for classifying the plant leaf. The process of deciding on the subset for all relevant features to be used in the construction of a system is known as feature selection. The Group Search Optimizer (GSO) is a nature-inspired algorithm that possesses all the qualities used effectively to solve feature selection tasks. In this work, there is a GSO-based algorithm of feature selection along with fuzzy logic and the classifier of a Neural Network (NN) is proposed. The results of the experiment prove the proposed method (GSO-NN) was able to achieve a better level of performance compared to the other methods.

RAPD-ANALYSIS OF CYCLAMEN SPP. GENOME POLYMORPHISM

Wild plants form the basis of biological resources both for Georgia and the whole world. A strategic task of any country is to preserve the biological diversity of plants. In the territory of Ajara, a large species diversity of plants grows, among which there are rare, endemic and relict plants. In particular, Cyclamen adzharicum.
 Modern systematics of wild plants in Georgia is based on classical methods of botany. In this regard, it is relevant to conduct genetic studies of species diversity and genetic polymorphism of species and populations using molecular genetic markers, in particular RAPD-PCR methods. The purpose of this study was to identify genetic polymorphism in Cyclamen L species using RAPD methods. As a result of the conducted research the 65 RAPD-markers in length from 150 to 1500 BP have been revealed. The number of the amplified fragments DNA varied depending on the primer from 6 (OPA-2) to 11 (OPB-4). The results of grouping Cyclamen adzharicum and C. coum samples allowed two clusters to be identified. In the first cluster were samples of three populations Cyclamen adzharicum and showed a low stubble in the intra-species variability. Cyclamen coum was attributed to the second cluster. The used primers gave the opportunity to identify polymorphism between the tested types of cyclamen.

Identification of Plant Leaf Disease using Machine Learning Techniques

Plant disease identification and classification is major area of research as majority of people in India depend on agriculture for their main source of income and for food. Identification of the diseases in any crops is challenging since manual identification techniques being used in this are based on the experts advises which may not be efficient. Based on leaf features decisions about variety of diseases are taken. In this paper an automated framework is introduced which can be used to detect and classify the diseases in the leaf accurately. Leaf images are acquired by using digital camera. Pre-processing techniques, segmentation and feature extraction are performed on the acquired images. The features are passed on to the classifiers to classify the diseases. This work has been proposed to classify and distinguish the leaf sample based on its features. The proposed work is carried out with Artificial Neural Network (ANN), Support Vector Machine (SVM) and Naive Bayes classifiers to analyze the result. For given dataset ANN performed better than the other two classifiers.

Targeted amplicon sequencing of 40 nuclear genes supports a single introduction and rapid radiation of Hawaiian Metrosideros (Myrtaceae)

Compared to traditionally used plastid or ribosomal markers, nuclear gene markers provide many advantages for molecular systematics of plants, and high-throughput sequencing technologies are making large nuclear datasets available at an unprecedented rate. We used targeted amplicon sequencing of 44 nuclear genes to construct a time-calibrated phylogeny of genus Metrosideros (Myrtaceae), evaluate recent systematic revisions, and assess whether phylogenetic signal within the Hawaiian Archipelago is correlated with island biogeography or morphological diversification. We generated a final dataset of 40 nuclear genes for 187 specimens, used multiple search heuristics and species-tree analysis to estimate a phylogeny, and incorporated new fossils for the genus to estimate divergence times across the dataset. All analyses supported the monophyly of Metrosideros, including Carpolepis and Tepualia. Hawaiian Metrosideros were monophyletic and dated to 3.1 MYA using new fossils for the genus, which is intermediate to previous estimates based on nuclear ribosomal/chloroplast loci and calibrated with island ages. Within the Hawaiian Metrosideros clade, we observed short branch lengths and unresolved relationships, and phylogenetic patterns were not concordant with biogeographic hypotheses of island progression, or the delineation of taxa or morphotypes. Average nucleotide diversity was relatively consistent across the Hawaiian Islands with the exception of slightly lower diversity on Kauaʻi. These results provide a data-rich estimate of the timing of a single introduction of Metrosideros to Hawai‘i and highlight the need for molecular markers with higher evolutionary rates for resolution of relationships within this recent radiation.

Weed Management of the Future

The methods used to protect agricultural products currently undergo drastic changes. Artificial Intelligence is a prime candidate to overcome two challenges faced by farmers around the world: The increasing cost and decreasing availability of human labor for weed control, and the growing global restriction of herbicides. Deep Learning is one of the most prominent approaches for applying AI to all kinds of use cases in industrial applications, entertainment, and security. Its latest field of application is plant classification that enables automated weed control and precise spot spraying of herbicides. While cheap, powerful platforms for deploying classification mechanisms are widely available, this comes at the cost of expensive and effort rich classifier training. This effectively makes Deep Learning-based approaches unavailable for the majority of the agricultural sector. Deepfield Robotics presents a systematic approach for deploying AI onto fields at large, including the learnings that led to their self-contained AI driven plant classification modules that relieve individuals from having to deploy their own AI solution. The same technology acts as enabler for more agricultural domains, such as targeted fertilization, nano irrigation, and automated phenotyping. This article documents Deepfield Robotics’ findings and vision on how AI can be the workhorse for agricultural weeding labor.

The criterion of conjunction in plant systematics and evolution

I review the theory and practice behind as reported by Patterson (in: Joysey, Friday (eds) Problems of phylogenetic reconstruction, Academic Press, London, 1982) the criterion of conjunction in plant systematics and evolution, with a focus on: (1) de Pinna (Cladistics 7:367–394, 1991) analysis of homology in the cladistic framework; (2) Hawkins’ (in: Scotland, Pennington (eds) Homology and systematics: coding characters for phylogenetic analysis, Taylor and Francis, London, 2000) survey of character coding; (3) Sereno (Cladistics 23:565–587, 2007) view of neomorphic and transformational characters; (4) character coding and polymorphic taxa; and (5) the relationship between character coding and plant variation using examples cited by Hawkins (in: Scotland, Pennington (eds) Homology and systematics: coding characters for phylogenetic analysis, Taylor and Francis, London, 2000). I coin the term “Replicable homology,” in contrast to serial homology, to make reference to the presence of multiple copies of the same structure or part in the same organism. I conclude that by Patterson’s (in: Joysey, Friday (eds) Problems of phylogenetic reconstruction, Academic Press, London, 1982) criterion is an important tool in order to identify neomorphic characters and that it cannot be applied to transformational characters. Conventional coding is the appropriate way to code characters, whereas both conjunction and unifying coding should be abolished from character analysis, as they are in disagreement with the view that a single character state must contain a mutually exclusive condition in relation to other character states.

Selection of interspecific <i>Psidium</i> spp. hybrids resistant to <i>Meloidogyne enterolobii

Since 2001, the disease known as 'guava decline', resulting from the interaction between the phytonematode Meloidogyne enterolobii and the fungus Fusarium solanie, has caused direct and indirect economic losses to the entire guava production chain. Given the lack of sources of resistance in guava genotypes, interspecific hybrids of Psidium spp. were obtained for resistance to the nematode M. enterolobii. To classify the level of resistance of the interspecific hybrids, we evaluated the plant classification methodologies proposed by Oostenbrink (1966) and Moura and Régis (1987). Estimates of genetic parameters were obtained using the REML/BLUP approach. Interspecific hybrids resistant to M. enterolobii were selected that can be used as rootstocks or in new crosses for the development of the guava breeding program.

Automated Leaf Alignment and Partial Shape Feature Extraction for Plant Leaf Classification

The last few decades have witnessed various approaches to automate the process of plant classification using the characteristics of the leaf. Several approaches have been proposed, and the majority focused on global shape features. However, one challenge that faces this task is the high interclass similarity amongst the leaves of different species in terms of the global shape. Furthermore, there always has been an obstacle against full automation as several approaches require user intervention to align the leaf. Therefore, a new set of Quartile Features (QF) is proposed in this paper to describe the partial shape of the leaf, in addition to an automated alignment approach to automate the system. The QF are extracted from the horizontal and vertical leaf quartiles to describe the partial shape of the leaf and the relations among its parts. The well-known Flavia dataset has been selected for the evaluation of the proposed system. The experimental results indicate the ability of the proposed alignment algorithm to align leaves with different shapes and maintain a correct classification accuracy regardless of the orientation of the input leaf samples. Furthermore, the proposed QF indicated promising results by increasing the accuracy of the classification by a range of approximately 26% to 30% when combined with Hu’s Moment Invariants, using k-fold cross-validation technique.

Hyb-Seq for Flowering Plant Systematics.

High-throughput DNA sequencing (HTS) presents great opportunities for plant systematics, yet genomic complexity needs to be reduced for HTS to be effectively applied. We highlight Hyb-Seq as a promising approach, especially in light of the recent development of probes enriching 353 low-copy nuclear genes from any flowering plant taxon.

Machine Learning-Based Spectral Library for Crop Classification and Status Monitoring

The establishment and application of a spectral library is a critical step in the standardization and automation of remote sensing interpretation and mapping. Currently, most spectral libraries are designed to support the classification of land cover types, whereas few are dedicated to agricultural remote sensing monitoring. Here, we gathered spectral observation data on plants in multiple experimental scenarios into a spectral database to investigate methods for crop classification (16 crop species) and status monitoring (tea plant and rice growth). We proposed a set of screening methods for spectral features related to plant classification and status monitoring (band reflectance, vegetation index, spectral differentiation, spectral continuum characteristics) that are based on ISODATA and JM distance. Next, we investigated the performance of different machine learning classifiers in the spectral library application, including K-nearest neighbor (KNN), Random Forest (RF), and a genetic algorithm coupled with a support vector machine (GA-SVM). The optimal combination of spectral features and the classifier with the highest classification accuracy were selected for crop classification and status monitoring scenarios. The GA-SVM classifier performed the best, which produced an accuracy of OAA = 0.94, Kappa = 0.93 for crop classification in a complex scenario (crops mixed with 71 non-crop plant species), and promising accuracies for tea plant growth monitoring (OAA = 0.98, Kappa = 0.97) and rice growth stage monitoring (OAA = 0.92, Kappa = 0.90). Therefore, the establishment of a plant spectral library combined with relevant feature extraction and a classification algorithm effectively supports agricultural monitoring by remote sensing.