House Plant Leaf Disease Detection and Classification Using Machine Learning

Hibiscus is a fantastic herb, and in Ayurveda, it is one of the most renowned herbs that have extraordinary healing properties. Hibiscus is rich in vitamin C, flavonoids, amino acids, mucilage fiber, moisture content, and antioxidants. Hibiscus can help with weight loss, cancer treatment, bacterial infections, fever, high blood pressure, lower body temperature, treat heart and nerve diseases. Automatic leaf disease detection is an essential task. Image processing is one of the popular techniques for the plant leaf disease detection and categorization. In this chapter, the diseased leaf is identified by concurrent k-means clustering algorithm and then features are extracted. Finally, reweighted KNN linear classification algorithms have been used to detect the diseased leaves categories.

Microbial Diversity Analysis and Genome Sequencing Identify Xanthomonas perforans as the Pathogen of Bacterial Leaf Canker of Water Spinach (Ipomoea aquatic)

Ipomoea aquatica is a leafy vegetable widely cultivated in tropical Asia, Africa, and Oceania. Bacterial leaf canker disease has been attacking the planting fields and seriously affecting the quality of I. aquatica in epidemic areas in China. This study examined the microbial composition of I. aquatica leaves with classical symptoms of spot disease. The results showed that Xanthomonas was overwhelmingly dominant in all four diseased leaf samples but rarely present in rhizospheric soil or irrigation water samples. In addition, Pantoea was also detected in two of the diseased leaf samples. Pathogen isolation, identification, and inoculation revealed that both Xanthomonas sp. TC2-1 and P. ananatis were pathogenic to the leaves of I. aquatic, causing crater-shaped ulcerative spots and yellowing with big brown rot lesions on leaves, respectively. We further sequenced the whole genome of strain TC2-1 and showed that it is a member of X. perforans. Overall, this study identified X. perforans as the causal pathogen of I. aquatica bacterial leaf canker, and P. ananatis as a companion pathogen causing yellowing and brown rot on leaves. The correct identification of the pathogens will provide important basis for future efforts to formulate targeted application strategy for bacterial disease control.

Biological and molecular characterization of seven Diaporthe species associated with kiwifruit shoot blight and leaf spot in China

Diaporthe species are significant pathogens, saprobes, and endophytes, with comprehensive host association and geographic distribution. These fungi cause severe dieback, cankers, leaf spots, blights, and stem-end rot of fruits on different plant hosts. This study, explored the occurrence, diversity and pathogenicity of Diaporthe spp. associated with Actinidia chinensis and A. deliciosa in the main kiwifruit production areas of China. Diaporthe isolates (284) derived from 106 diseased leaf and branch samples were examined. Multi-locus phylogenetic analyses and morphology of 43 representative isolates revealed that seven Diaporthe species were obtained, including D. alangii, D. compactum, D. eres, D. hongkongensis, D. sojae, D. tectonae, and D. unshiuensis. Pathogenicity tests were performed on kiwifruit fruits, leaves and branches. Koch’s postulates confirmed all species were pathogenic. D. alangii and D. tectonae were the most aggressive species, followed by D. eres, D. sojae, D. hongkongensis, D. unshiuensis, and D. compactum. Host range evaluation showed that the seven Diaporthe species could also infect apricot, apple, peach, pear, and plum. This is the first report of D. alangii, D. compactum, D. sojae, D. tectonae, and D. unshiuensis infecting kiwifruit in China, increasing understanding of the Diaporthe complex causing diseases of kiwifruit plants, to assist effective disease management.

A 3-Stage Method for Disease Detection of Cotton Plant Leaf using Deep Learning CNN Algorithm

Agriculture is one of the significant occupation in various countries including India. As major part of the Indian financial system is reliant on agriculture production, the intense consideration to the concern of food production is essential. The nomenclature and recognition of crop infection got much significance in technical as well as economic in the Agricultural Industry. While keeping track of diseases in plants with the support of experts can be very expensive in agriculture region. There is a necessity for a method or system which can automatically identify the diseases as it can bring revolution in monitoring enormous fields of crop and then plant leaflet can be taken ca The detection of cotton leaf disease is a very important factor to prevent serious outbreak.re imme4diately after recognition of disease. The aim of this paper is to provide guidelines for the development of application which recognizes cotton plant leaf diseases. For availing this user need to upload the image of the cotton leaf and then with the help of image processing one can get a digitized colour image of a diseased leaf which can be further processed by applying CNN algorithm to predict the actual root cause for the cotton leaf disease.

First Report of Dickeya fangzhongdai causing soft rot in Orchids in Canada

Dickeya fangzhongdai was originally described as the causal agent of bleeding canker of pear tree in China. Recently, D. fangzhongdai was isolated and identified as the causal agent of soft rot in an orchid plant purchased in a local supermarket in Prince Edward Island, Canada. A water-soaked dark green spot on the leaf surface was observed and later became larger soft rot symptom. The origin of the orchid plants was traced back to a producer in Ontario, Canada who propagated them from with cuttings originally imported from the Netherlands and Taiwan. Bacterial isolations were made from a soft rot lesion on an orchid leaf by surface sterilization of small pieces of marginal tissue of the diseased leaf in 70% alcohol. The small pieces of leaf tissue were then washed three time using sterile water, and immersed in drops of sterile water. Bacterial streaming was observed under the microscope and non-fluorescing bacterial colonies were isolated on King’s B and casamino acid-peptone-glucose agar plates and purified as isolates 908, 909, 910 and 911. The DNA samples were extracted from the four isolates, as well as the diseased leaf tissue, and tested by using a qPCR assay with the specific primer/probe set (DfF/DfR/DfP) for D. fangzhongdai (Tian et al. 2020). The assay yielded PCR amplicons of 135 bp with a melting temperature of 86.5±0.6 °C as did two control reactions using genomic DNA from D. fangzhongdai strains JS5T and QZH3 originally isolated in China, providing presumptive identification of the orchid isolates as D. fangzhongdai. To fulfill Koch’s postulates, freshly purchased healthy orchid plants (n=4) were inoculated by leaf injection with the bacterial isolates obtained in this study and strains JS5 T and QZH3 at ~107 CFU/ml. Three leaves of the same side of the plants were inoculated with the same strains as triplicates. Sterile water was used as the negative control. Inoculated plants were incubated in a growth chamber with a 16 h photoperiod at 23 °C. Water soaked lesions developed in 3-5 days after inoculation followed by soft rotting in leaves inoculated with the new bacterial strains from orchid plants while strain QZH3 caused soft rot in 10 days after inoculation (Fig. S1). The non-fluorescing bacteria on King’s B plates with colony morphology similar to those inoculated were re-isolated from the inoculated leaves and confirmed to be D. fangzhongdai by qPCR. Phylogenetic analysis of the assembled 16S rRNA sequence of isolate 908 (GenBank accession number: MT984340), together with GenBank data of all Dickeya spp. and some Pectobacterium spp, using neighbor-joining (NJ) method inferred with MEGA X software (Kumar et al. 2018) showed that isolate 908 clustered with strains JS5T and QZH3 at a phylogenetic distance of 0.0007. This clearly indicated that isolate 908 and JS5T and QZH3 belong to the same genus. Species-level identification of isolate 908 was achieved by genome sequencing and analysis based on average nucleotide identity (ANI). Genomic DNA of isolate 908 was sequenced with Illumina MiSeq to provide approximately 180X genome coverage. After quality checking using FastQC (Andrews 2010), de-novo assembly was performed with VelvetOptimiser v2.2.6 (Zerbino and Birney 2008). The draft genome size of strain 908 was 4,938,027 bp consisting of 76 contigs with 56.8% G+C content and 63,801 bp as N50. The draft genome was checked for misassembled fragments using QUAST v5.0.2 (Gurevich et al. 2013) and found to be of good quality. The draft genome sequence is deposited in GenBank under the accession number of JADCNJ000000000. The draft genome sequence of strain 908 was compared to that of D. fangzhongdai JS5T type strain genome using FastANI v1.2 (Jain et al. 2018) resulting in an ANI value of 98.9%, which is above the 95% cut-off for the same species. Previously, it was reported that D. fangzhongdai caused soft rot in orchid in Europe (Alič et al. 2018) and in onions in New York (Ma et al. 2020). The difference in virulence among D. fangzhongdai strains warrants further investigation and their pathogenicity on potato is being investigated to evaluate any threat to the potato industry. To our knowledge, this is the first report of D. fangzhongdai causing soft rot disease on orchids in Canada and North America.

Biochemical Changes in the leaf galls of Cinnamomum Verum Presl. (Syn. Cinnamomum zeylanicum) induced by unknown Eriophyes sp.

The genus Cinnamomum is a member of a tropical evergreen tree of the Lauraceae family. In its wild state, it grows up to 7m (56 ft). Cinnamomum verum Presl. (Syn. Cinnamomum zeylanicum) is of commercial value and is extensively used in culinary as a spice in food and the ayurvedic system of medicine. One of the major diseases, causing severe losses in yield, thus affecting the economy in India is the leaf gall disease of Cinnamon. To understand the host-pathogen interactions, it becomes obligatory to estimate the proteins, carbohydrates, enzymes, etc. present in a particular host plant quantitatively, to draw meaningful conclusions on host-pathogen interaction. The pathogen is always associated with the infection caused to a healthy plant. Eventually, biochemical changes take place in the diseased tissue. The pathogenic organism releases cell secretion, which comprises of various cell metabolites which alter the metabolism of the diseased tissue. In the present investigation, changes in the biochemical profile of healthy and diseased leaf of Cinnamon has been attempted, and the results have been discussed in the light of pathogenicity, induced by unknown Eriophyes sp.

Plant Disease Detection and Classification Using Image Processing and Neural Networks

Plant disease is a major problem for food security, but in many parts of the world their rapid prediction remains difficult because of lack of the infrastructure required. New advances in machine vision achieved via deep learning have paved the way for diagnosis of AI-assisted diseases. To help determine the extent of plant disease, provide agricultural specialists with a digital archive with photos of diseased leaves. Estimate depends on Standard Area Diagrams (SADs), a collection of diseased leaf images, each of which includes an incrementally more diseased leaf compared to the previous one. Every SAD shows seriousness of the disease in terms of the percentage of the diseased leaf. Users then turn to the field for a leaf. For eg, equate it to SADs and use it to measure the severity of the disease. “This app is useful for crop consultants and research scientists looking to cut costs and improve the time and accuracy for assessing disease severity in plants.”

First Report of Black Spot Caused by Neoscytalidium dimidiatum on Sisal in Guangxi, China

Sisal (Agave sisalana Perrine) is an important hard fiber crop that is widely planted in Guangxi, Guangdong, Hainan, Yunnan, and Fujian provinces, China. In July 2019, a new leaf disease of sisal with a disease incident of about 36% was found in Guangxi (Fig.1a~d). The oval or circular black lesions were 2.3 cm to 15.9 cm in length and 1.6 cm to 5.5 cm in width on both sides of the diseased leaves. The central part of the lesions was slightly hollow. The lesions continuously enlarged and ultimately penetrated the leaves. Reddish brown and dark mucus was secreted from the lesions. The junction of lesions and healthy parts was reddish brown to yellow. The diseased leaf fiber and mesophyll tissues were reddish brown and necrotic. Fresh leaf yield was reduced about 30% by the disease, and fiber quality was significantly compromised every year in Guangxi. Six kinds of fungi distinguished by their morphology, size and color of the colonies were isolated from diseased leaf tissues of 60 sisal plants sampled from five different farms in Guangxi. Isolate JMHB1 was isolated at a rate of 95.67%. The isolate JMHB1 was initially white with dense and hairy aerial mycelium, gradually turning dark grey to olive green on PDA (Fig. 2). Conidia, arthrospores, and chlamydospores were observed on PDA in culture (Fig. 3). The conidia formed arthric chains, disarticulating, cylindrical-truncate, oblong-obtuse to doliiform, colorless and transparent, zero- to one-septate, and averaging 4.4 to 13.8 µm × 2.2 to 5.6 µm (n=100). Arthrospores were short columnar, pigmented and transparent, single or formed arthric chains, averaging 5.5 to 17.9 µm × 2.1 to 3.5 µm (n=100). Chlamydospores were dark brown, round or oval, averaging 4.5 to 9.6 µm × 4.5 to 8.6 µm (n=100). Pathogenicity testing was conducted by inoculating 3-year-old healthy sisal plants with PDA plugs (5 × 5 mm) on which the fungus had grown for 5 days. Nine healthy plants were wounded on the leaves with a sterile needle, and mycelial plugs were placed on the wounds, covered with sterile moist cotton, and wrapped with parafilm. Nine control plants were wounded and treated with PDA plugs as the negative control. The test was repeated three times. All treated plants were kept in a greenhouse at ~28 ℃ and 40% RH. After 5 days, only leaves inoculated with isolate JMHB1 showed lesions similar to symptoms observed in the field (Fig.1e~f). The fungus was re-isolated from all nine diseased plants, and no symptoms were observed on the leaves of control plants. Molecular identification of the fungus was made by PCR amplification of the internal transcribed spacer (ITS) region of rDNA, EF1-α gene and β-tubulin gene using primers ITS1/ITS4 (White et al. 1990), EFl-728F/EF1-986R (Carbone and Kohn 1999), TUB2Fd/TUB4Rd (Aveskamp et al. 2009) respectively. The ITS (MT705646), EF1-α (MT733516) and β-tubulin (MT773603) sequences of JMHB1 were similar to the ITS (AY819727), EF1-α (EU144063) and β-tubulin (KF531800) sequences of the epitype of Neoscytalidium dimidiatum (CBS 499.66) with 100%, 99.65% and 99.02% identity, respectively. Based on pathogenicity testing, morphological characteristics, and molecular identification, the pathogen of sisal causing black spot was identified as N. dimidiatum (Penz.) Crous & Slippers (Crous et al. 2006). To our knowledge, this is the first report of black spot caused by N. dimidiatum on sisal in China. Sisal is the main economic crop in arid and semi-arid areas that is widely planted in several provinces of southern China. The serious occurrence of the disease caused by N. dimidiatum has greatly affected the development of sisal industry and local economic income in China. Identification of the pathogen of the disease is of great significance to guide disease control, increase farmers' income and promote the development of sisal industry. References: Aveskamp, M. M., et al. 2009. Mycologia, 101: 363. https://doi.org/10.3852/08-199. Carbone, I., and Kohn, L. M. 1999. Mycologia, 91:553. https://doi.org/10.1080/00275514.1999. 12061051. Crous, P. W., et al. 2006. Stud. Mycol. 55:235. https://doi.org/10.3114/sim.55.1.235. White, T. J., et al. 1990. PCR Protocols: A Guide to Methods and Applications. Academic Press, San Diego, Page 315. doi.org/10.1002/mrd.1080280418. Supplemental photographs: Fig. 1 Symptoms of sisal black spot disease a, b, c, d showed symptoms in the field, e and f were symptoms after inoculating Neoscytalidium dimidiatum JMHB1. a, c, and e were the front of the lesions, b, d, and f were the back of the lesions. Fig. 2 Primary colony (a) and old colony (b) of Neoscytalidium dimidiatum JMHB1 Fig. 3 Arthrospores (a), conidia and chlamydospores (b) of Neoscytalidium dimidiatum JMHB1

Field Characterization of Partial Resistance to Gray Leaf Spot in Elite Maize Germplasm

Forty-eight inbred lines of maize with varying levels of resistance to gray leaf spot (GLS) were artificially inoculated with Cercospora zeina and evaluated to characterize partial disease resistance in maize under field conditions from 2012 to 2014 across 12 environments in western Kenya. Eight measures of disease epidemic—that is, final percent diseased leaf area (FPDLA), standardized area under the disease progress curve (SAUDPC), weighted mean absolute rate of disease increase (ρ), disease severity scale (CDSG), percent diseased leaf area at the inflection point (PDLAIP), SAUDPC at the inflection point (SAUDPCIP), time from inoculation to transition of disease progress from the increasing to the decreasing phase of epidemic increase (TIP), and latent period (LP)—were examined. Inbred lines significantly (P < 0.05) affected all measures of disease epidemic except ρ. However, the proportion of the variation attributed to the analysis of variance model was most strongly associated with SAUDPC (R2 = 89.4%). Inbred lines were also most consistently ranked for disease resistance based on SAUDPC. Although SAUDPC was deemed the most useful variable for quantifying partial resistance in the test genotypes, the proportion of the variation in SAUDPC in each plot was most strongly (R2 = 93.9%) explained by disease ratings taken between the VT and R4 stages of plant development. Individual disease ratings at the R4 stage of plant development were nearly as effective as SAUDPC in discerning the differential reaction of test genotypes. Thus, GLS rankings of inbred lines based on disease ratings at these plant developmental stages should be useful in prebreeding nurseries and preliminary evaluation trials involving large germplasm populations.