Tea () plantations are exposed to biotic and abiotic stresses. Among the biotic factors, blister blight (BB), caused by , affects the quality and quantity of the product and demands high fungicide application. A long term solution for disease resistance would require the knowledge of the basic molecular and biochemical changes occurring in plant as an attempt to resist the pathogen and limit the spread of the disease which can further help in developing resistant cultivars using biotechnological tools. Thus, gene expression studies using the cDNA based suppressive subtractive hybridization library, characterization of genes for pathogenesis related (PR) proteins [chitinase (), glucanase (), phenylalanine ammonia lyase ()] and genes in flavonoid pathway were accessed in the BB resistant and susceptible cultivars, SA6 and TES34, respectively. Further, biochemical analysis of PR and antioxidant enzymes (POX, APX, SOD) involved in BB resistance have been carried out to investigate the potential molecular and biochemical changes. Various stages of pathogen development had varied impact on PR protein, flavonoid pathway and anti-oxidative enzymes and indicates the possible role of reactive oxygen species, lignins, flavonoids, anthocyanins and other synthesized compounds in acting as antimicrobial/antifungal agents in tea cultivars.

We proposed a novel deep convolutional neural network (DCNN) using inverted residuals and linear bottleneck layers for diagnosing grey blight disease on tea leaves. The proposed DCNN consists of three bottleneck blocks, two pairs of convolutional (Conv) layers, and three dense layers. The bottleneck blocks contain depthwise, standard, and linear convolution layers. A single-lens reflex digital image camera was used to collect 1320 images of tea leaves from the North Bengal region of India for preparing the tea grey blight disease dataset. The nongrey blight diseased tea leaf images in the dataset were categorized into two subclasses, such as healthy and other diseased leaves. Image transformation techniques such as principal component analysis (PCA) color, random rotations, random shifts, random flips, resizing, and rescaling were used to generate augmented images of tea leaves. The augmentation techniques enhanced the dataset size from 1320 images to 5280 images. The proposed DCNN model was trained and validated on 5016 images of healthy, grey blight infected, and other diseased tea leaves. The classification performance of the proposed and existing state-of-the-art techniques were tested using 264 tea leaf images. Classification accuracy, precision, recall, F measure, and misclassification rates of the proposed DCNN are 98.99%, 98.51%, 98.48%, 98.49%, and 1.01%, respectively, on test data. The test results show that the proposed DCNN model performed superior to the existing techniques for tea grey blight disease detection.

No abstract