A lightweight convolutional neural network for recognition of severity stages of maydis leaf blight disease of maize.

Maydis leaf blight (MLB) of maize (Zea Mays L.), a serious fungal disease, is capable of causing up to 70% damage to the crop under severe conditions. Severity of diseases is considered as one of the important factors for proper crop management and overall crop yield. Therefore, it is quite essential to identify the disease at the earliest possible stage to overcome the yield loss. In this study, we created an image database of maize crop, MDSD (Maydis leaf blight Disease Severity Dataset), containing 1,760 digital images of MLB disease, collected from different agricultural fields and categorized into four groups viz. healthy, low, medium and high severity stages. Next, we proposed a lightweight convolutional neural network (CNN) to identify the severity stages of MLB disease. The proposed network is a simple CNN framework augmented with two modified Inception modules, making it a lightweight and efficient multi-scale feature extractor. The proposed network reported approx. 99.13% classification accuracy with the f1-score of 98.97% on the test images of MDSD. Furthermore, the class-wise accuracy levels were 100% for healthy samples, 98% for low severity samples and 99% for the medium and high severity samples. In addition to that, our network significantly outperforms the popular pretrained models, viz. VGG16, VGG19, InceptionV3, ResNet50, Xception, MobileNetV2, DenseNet121 and NASNetMobile for the MDSD image database. The experimental findings revealed that our proposed lightweight network is excellent in identifying the images of severity stages of MLB disease despite complicated background conditions.

Population Structure Analysis and Association Mapping for Turcicum Leaf Blight Resistance in Tropical Maize Using SSR Markers.

Maize is an important cereal crop in the world for feed, food, fodder, and raw materials of industries. Turcicum leaf blight (TLB) is a major foliar disease that can cause more than 50% yield losses in maize. Considering this, the molecular diversity, population structure, and genome-wide association study (GWAS) for TLB resistance were studied in 288 diverse inbred lines genotyped using 89 polymorphic simple sequence repeats (SSR) markers. These lines werescreened for TLB disease at two hot-spot locations under artificially inoculated conditions. The average percent disease incidence (PDI) calculated for each genotype ranged from 17 (UMI 1201) to 78% (IML 12-22) with an overall mean of 40%. The numbers of alleles detected at a locus ranged from twoto nine, with a total of 388 alleles. The polymorphic information content (PIC) of each marker ranged between 0.04 and 0.86. Out of 89 markers, 47 markers were highly polymorphic (PIC ≥ 0.60). This indicated that the SSR markers used were very informative and suitable for genetic diversity, population structure, and marker-trait association studies.The overall observed homozygosity for highly polymorphic markers was 0.98, which indicated that lines used were genetically pure. Neighbor-joining clustering, factorial analysis, and population structure studies clustered the 288 lines into 3-5 groups. The patterns of grouping were in agreement with the origin and pedigree records of the genotypesto a greater extent.A total of 94.10% lines were successfully assigned to one or another group at a membership probability of ≥0.60. An analysis of molecular variance (AMOVA) revealed highly significant differences among populations and within individuals. Linkage disequilibrium for r2 and D' between loci ranged from 0 to 0.77 and 0 to 1, respectively. A marker trait association analysis carried out using a general linear model (GLM) and mixed linear model (MLM), identified 15 SSRs markers significantly associated with TLB resistance.These 15 markers were located on almost all chromosomes (Chr) except 7, 8, and 9. The phenotypic variation explained by these loci ranged from 6% (umc1367) to 26% (nc130, phi085). Maximum 7 associated markers were located together on Chr 2 and 5. The selected regions identified on Chr 2 and 5 corroborated the previous studies carried out in the Indian maize germplasm. Further, 11 candidate genes were identified to be associated with significant markers. The identified sources for TLB resistance and associated markers may be utilized in molecular breeding for the development of suitable genotypes.

Deep learning-based approach for identification of diseases of maize crop.

In recent years, deep learning techniques have shown impressive performance in the field of identification of diseases of crops using digital images. In this work, a deep learning approach for identification of in-field diseased images of maize crop has been proposed. The images were captured from experimental fields of ICAR-IIMR, Ludhiana, India, targeted to three important diseases viz. Maydis Leaf Blight, Turcicum Leaf Blight and Banded Leaf and Sheath Blight in a non-destructive manner with varied backgrounds using digital cameras and smartphones. In order to solve the problem of class imbalance, artificial images were generated by rotation enhancement and brightness enhancement methods. In this study, three different architectures based on the framework of 'Inception-v3' network were trained with the collected diseased images of maize using baseline training approach. The best-performed model achieved an overall classification accuracy of 95.99% with average recall of 95.96% on the separate test dataset. Furthermore, we compared the performance of the best-performing model with some pre-trained state-of-the-art models and presented the comparative results in this manuscript. The results reported that best-performing model performed quite better than the pre-trained models. This demonstrates the applicability of baseline training approach of the proposed model for better feature extraction and learning. Overall performance analysis suggested that the best-performed model is efficient in recognizing diseases of maize from in-field images even with varied backgrounds.

Efficacy of Newer Molecules, Bioagents and Botanicals against Maydis Leaf Blight and Banded Leaf and Sheath Blight of Maize.

Maize (Zea mays L.; 2N=20) is major staple food crop grown worldwide adapted to several biotic and abiotic stresses. Maydis leaf blight (MLB) and banded leaf and sheath blight (BLSB) are serious foliar fungal diseases may cause up to 40% and 100% grain yield loss, respectively. The present studies were undertaken to work out the efficacy of chemicals, botanicals and bioagents for the management of MLB and BLSB under field condition for two seasons Kharif 2014 and 2015. Five molecules (propiconazole 25 EC, hexaconazole 25 EC, carbendazim 50 WP, mancozeb 75 WP and carbedazim 12 WP + mancozeb 63 WP), two bioagents i.e. Trichoderma harzianum and T. viridae and three botanicals namely azadirachtin, sarpagandha and bel pathar were tested for their efficacy against MLB. Eight newer fungicides viz., difenconazole 250 SC, hexaconazole 5 EC, carbendazim 50WP, validamycin 3 L, tebuconazole 250 EC, trifloxystrobin 50 WG + tebuconazole 50 WG, azoxystrobin 250 EC and pencycuron 250 SC were evaluated against BLSB. Analysis revealed significant effects of propiconazole at 0.1%, carbendazim 12 WP + mancozeb 63 WP at 0.125% and sarpagandha leaves at 10% against MLB pathogen, whereas validamycin at 0.1% and trifloxystrobin 25 WG + tebuconazole 50 WG at 0.05% were found effective against BLSB. The slow rate of disease control virtually by the bioagents might have not shown instant effect on plant response to the yield enhancing components. The identified sources of management can be used further in strengthening the plant protection in maize against MLB and BLSB.