• Journal of Multimedia Information System
• /
• v.6 no.2
• /
• pp.49-60
• /
• 2019

Plants are very crucial for life on Earth. There is a wide variety of plant species available, and the number is increasing every year. Species knowledge is a necessity of various groups of society like foresters, farmers, environmentalists, educators for different work areas. This makes species identification an interdisciplinary interest. This, however, requires expert knowledge and becomes a tedious and challenging task for the non-experts who have very little or no knowledge of the typical botanical terms. However, the advancements in the fields of machine learning and computer vision can help make this task comparatively easier. There is still not a system so developed that can identify all the plant species, but some efforts have been made. In this study, we also have made such an attempt. Plant identification usually involves four steps, i.e. image acquisition, pre-processing, feature extraction, and classification. In this study, images from Swedish leaf dataset have been used, which contains 1,125 images of 15 different species. This is followed by pre-processing using Gaussian filtering mechanism and then texture and color features have been extracted. Finally, classification has been done using Multiclass-support vector machine, which achieved accuracy of nearly 93.26%, which we aim to enhance further.

• Plant Resources
• /
• v.8 no.3
• /
• pp.244-249
• /
• 2005

The growth charateristics and karyotypes of Aster spathulifolius collected from 5 sites including coastal and island region on the Korean peninsula, were analysed. Several morphological characteristics of the plants such as leaf length, leaf width, top internode, medium internode, spike branching, flower diameter, number of petal, leaf color, leaf form, stem and leaf hair, viscosity, and serration of the plants were distinctly different depending on the native region from which they were collected. Karyotypic analysis showed that the chromosome number was all diploid (2n=18), with one pair of submetacentric satellite chromosomes. The chromosome composition included 7 pairs of metacentric chromosomes and 2 pairs of submetacentric chromosomes in all plants. However, chromosome order and the ranges of the chromosome lengths were a little different from plant to plant according to their native growing regions. The plants from Geoje-Do especially showed large differences in the chromosome lengths between the longest and the shortest compared to the plants from other places. This results provide important data to support the classification of the species into several sub-species.

• Korean Journal of Plant Resources
• /
• v.19 no.1
• /
• pp.105-111
• /
• 2006

The objectives of this study, an analysis of fruit and leaf morphological characteristics among the five Zizyphus cultivars could be used for the investigation of cultivars classification and could provide information to make out the UPOV TG(Test Guidelines). ANOVA tests showed that there were statistically significant differences in all fruit and leaf morphological characteristics among the five Zizyphus cultivars at 1% level. But, for kernel characteristics, differences were statistically non-significant among the cultivars. Approximately, the Wolchul and Boeun cultivars showed larger and smaller values in overall characteristics and cultivars, respectively. The results of principal component analysis(PCA) for the fruit and leaf morphological characteristics showed that the first for principal components(PC's) explained about 65.3% of the total variation. The first PC was correlated with those characteristics that were mainly related to the terminal leaf length(TLL), leaf length(LL), fruit length(FL), terminal leaf width(TLW), and leaf petiole length(LPL). The second and third PC was mainly correlated with the terminal leaf morphological index(TLMI). Therefore, these characteristics were important to analysis of the fruit and leaf morphological characteristics and classification among the five Zizyphus cultivars. Cluster analysis using UPGMA method based on principal components showed that five Zizyphus cultivars could be clustered into two groups. Group I comprises Mudung, Wolchul, and Bokjo and Geumsung cultivars, Group II is Boeun cultivar. These results well similar to that of principal component analysis.

• Journal of the Korean Society of Tobacco Science
• /
• v.5 no.2
• /
• pp.25-32
• /
• 1983

To obtain basic information on the breeding of burley tobacco, classification of 41 varieties was carried out by using the cluster analysis of correlation coefficients and taxonomic distance based on twenty-one agromonic characters. Eight characters, such as days to flowering, length of flower axis, internode length, leaf length, yield, leaf angle to stem, vein angle to midrib and plant height, were useful in monothetic classification. Forty-one varieties were classified into four groups (I, II, III and IV) with weighted variable group method (WVGM ) and weighted jai. group method(WPGM), whereas the results classification of 33 varieties among them by WVGM were coincident with the results by WPGM. As for the characteristics of each group, group I related to late maturity, tall height and high yield, group II related to intermediate maturity, tall height and low yield, group 19 related to early maturity, intermediate height and low yield, and group W related to early maturity, short height and intermediate yield.

• Journal of Broadcast Engineering
• /
• v.25 no.2
• /
• pp.208-217
• /
• 2020

The classification of plant diseases by images captured by a camera sensor has been studied over past decades. A method that has gained much interest is to use image segmentation, from which statistical features are derived and analyzed by machine learning. Recently, deep learning has been adopted in this area. However, image segmentation is still a difficult task to achieve stable performance due to a variety of environmental variations. The end-to-end learning in neural network has a demerit that train images may be different from real images acquired in outdoor fields. To solve these problems, we propose superpixel-based disease classification method using end-to-end CNN (convolutional neural network) learning. Based on experiments performed on PlantVillage apple images, the classification accuracy is 98.29% and 92.43% for full-image and superpixel. As well, the multivariate F1-score is (0.98, 0.93). Therefore we validate that the method of using superpixel is comparable to that of full-image.

• Journal of Plant Biology
• /
• v.29 no.1
• /
• pp.19-40
• /
• 1986

A study on the species description, enumeration and interspecific relationships of genus Filipendula was undertaken for four Korean and two Japanese species. The use of scanning electron miroscope(SEM) has made possible a detailed study of foliar trichomes, pollen grains and fruit surface features. A new method for the preparation of SEM samples was developed in this study. Filipendula glaberrima, F. formosa, and F. korean were reliable endemic species. Among them, F. glaberrima shares many characteristics with F. formosa. F. gloaberrima can be divided into tow types based on leaf, seed, flower and pollen characters. Type I is characterized by ciliated fruits, many branched vascular bundles on the petal, dense trichomes on the leaf vein of abaxial surface, and many foveolate pollen surface. Type II possesses glabrous fruits, a few vascular bundles on the petal, rare trichomes on the leaf vein and a few foveolate pollen surface. The significant taxonomic characteristics in the classification of genus Filipendula are their lengths of fruit stalks, the densities of marginal trichomes of fruits, and textures of stipules.

• Journal of Korea Multimedia Society
• /
• v.17 no.3
• /
• pp.300-311
• /
• 2014

According to fast change of the environment, the structured study of the ecosystem by analyzing the plant leaves are needed. Expecially, the methodology that searches and classifies the leaves from captured from the smart device have received numerous concerns in the field of computer science and ecology. In this paper, we propose a plant leaf classification technique using shape descriptor by combining Scale Invarinat Feature Transform (SIFT) and Histogram of Oriented Gradient (HOG) from the image segmented from the background via Graphcut algorithm. The shape descriptor is coded in the field of Locality-constrained Linear Coding to optimize the meaningful features from a high degree of freedom. It is connected to Support Vector Machines (SVM) for efficient classification. The experimental results show that our proposed approach is very efficient to classify the leaves which have similar color, and shape.

• Proceedings of The Korean Society of Agricultural and Forest Meteorology Conference
• /
• 2001.06a
• /
• pp.93-96
• /
• 2001

Estimation of leaf wetness duration (LWD) facilitates assessment of the likelihood of outbreaks of many crop diseases. Models that estimate LWD may be more convenient and grower-friendly than measuring it with wetness sensors. Empirical models utilizing statistical procedures such as CART (Classification and Regression Tree; Gleason et al., 1994) have estimated LWD with accuracy comparable to that of electronic sensors.(omitted)

• Plant Resources
• /
• v.2 no.2
• /
• pp.75-80
• /
• 1999

Twenty two varieties of pecan including wild types were classified based on 6 characters measured by principal component analysis score distance. The results are summarized as fellow. Twenty two varieties were classified into 5 groups based in PCA score distance. Five groups were distinctly characterized by many morphological characters. Total variation could be explained by 51%, 95%, 99% with first, third and fifth principal components respectively. Varimax rotation of the factor loading of the first factors indicated that the first component was highly loaded with leaf characters, the second component with fruit characters, but fruit length was negative loaded. The second, the third and the fourths groups of cultivars had very close genetic parentage similarity.

• Journal of Plant Biology
• /
• v.10 no.1_2
• /
• pp.3-20
• /
• 1967

1. Comparative studies, the number, form and pattern of ramification, on the petiole stele types of 3 orders, 11 families, 41 genera and 104 species of ferns found in Korea were carried out. 2. The number, form and pattern of the ramified steles were found to be different according to the taxa studied. 3. The stele types of petiole may be classified as unibranch, bibranch, tribranch, and polybranch. The species belonging to each stele type were found to have similar embyrological characteristics among them. Therefore, it might be reasonable to assume that the stele type can be used as a basis for classifing family lines. 4. The number of ramified steles in the petiole were found to be in general agreement with that of the leaf traces, though a few exceptional cases were found. 5. It is well known that there is a large degree of disgreement among the taxanomists on the classification of ferns. The classification of ferns by means of petiole stele types may ease this difficulty in certain extent.