Apple Leaf Research Articles

Design and Implementation of Apple Leaf Disease Recognition System Based on ResNet50

Timely identification of apple leaf diseases is critical to preventing crop losses and safeguarding yields. Spotted leaf drop, brown spot, gray spot, mosaic and rust are all common types of apple leaf diseases, and their presence signals a potential risk that could lead to significant reductions in fruit and crop yields. The apple industry is thus at risk of economic losses. In order to solve this problem, this study applies ResNet50, a deep learning model for image recognition and classification of apple leaf diseases, and develops an intelligent recognition system for apple leaf diseases by combining PyQt technology. The purpose of this system is to overcome the shortcomings of traditional recognition methods. Through experimental validation, the ResNet50 model achieves a high accuracy rate of 93.19% in apple leaf disease recognition, demonstrating its efficiency and practicality in practical applications.

Identification the Pathogen Cause a New Apple Leaf Blight in China and Determination the Controlling Efficacy for Five Botanical Fungicides.

Alternaria leaf blight has recently been described as an emerging fungal disease of apple trees which is causing the significant damage in the apple-growing areas of Tianshui and Jingning, Gansu, China. In the present study, the pathogen species involved in apple leaf blight and its biological characteristics were identified, and the inhibitory activity of different botanical fungicides against the pathogen was evaluated in vitro. Four strains were isolated from the symptomatic areas of necrotic apple leaves, and initially healthy leaves showed similar symptoms to those observed in orchards after inoculation with the ABL2 isolate. The ABL2 isolate was identified as Alternaria tenuissima based on the morphological characteristics of its colonies, conidiophores, and conidia, and this was also confirmed by multi-gene sequence (ITS, OPA10-2, Alta-1, and endoPG) analysis and phylogenic analysis. The optimum temperature, pH, carbon source, and nitrogen source for the growth of A. tenuissima mycelia were 28 °C, 6-7, soluble starch, and soy flour, respectively. In addition, the botanical fungicide eugenol exhibited the highest inhibitory effect on the mycelial growth and conidia germination of A. tenuissima, and the median effective concentration (EC50) values were 0.826 and 0.755 μg/mL, respectively. The protective and curative efficacy of eugenol were 86.85% and 76.94% after inoculation in detached apple leaves at a concentration of 4 μg/mL. Our research provides new insights into the control of apple leaf blight disease by applying botanical fungicides.

Apple leaf diseases recognition based on ResNet-101 and CBAM

Early identification and management of plant diseases is crucial to raising crop quality and yield. In order to create an effective deep learning model for autonomous apple leaf diseases identification, the issue of recognizing and classifying apple leaf infections is the main focus of this investigation. To improve the detection performance of the model, Convolutional Block Attention Module (CBAM) is added to the ResNet-101 deep convolutional neural network (CNN) model utilized in this work. According to the experimental findings, the ResNet-101 model with Convolutional Block Attention Module exhibits excellent performance when attempting to diagnose problems in apple leaves. Specifically, the ResNet-101+CBAM model outperforms the standard ResNet-101 model in the disease categories of 'Brown spot', 'Grey spot', 'Health' and 'Rust'. In some disease categories, such as 'Powdery mildew', the performance enhancement of the module (CBAM) is relatively insignificant due to the confusion caused by the similarity of features between the categories. Further analysis showed that the data enhancement technique had a significant positive impact on model performance.

Apple Leaf Disease Prediction Using Deep Learning Models

In India, nearly two lakh people rely on apple production for a living, with apples accounting for approximately 16% of GDP in their respective states. Although apple production is second only to that of other fruits in terms of volume, it generates more revenue. In addition, India ranks fifth in total apple production worldwide. As a result, the apple is a fruit that has a significant impact on our economic situation. Apple productivity is primarily influenced by leaf diseases such as scab, cedar rust, black rot, and others. To prevent apple leaf diseases from spreading throughout the tree, it is critical to detect them in their early stages of infection. We proposed a convolutional neural network methodology to accurately and efficiently detect these leaf diseases. The image pre-processing and augmentation methods help us distinguish the region of interest from the background. We then trained and validated our dataset of apple leaf images by applying deep learning algorithms. The Dense Net algorithm, which has a 98.42% accuracy rate, is used to label the leaf diseases in our model.

A novel framework to assess apple leaf nitrogen content: Fusion of hyperspectral reflectance and phenology information through deep learning

Rapid and accurate assessment of apple-tree leaf nitrogen content (LNC) plays an important role in smart orchards for accurate fertilization. Hyperspectral remote sensing technology provides a reliable, economical, and fast method to estimate LNC. However, high-dimensional spectral data contain redundant information that is irrelevant to the target parameters, necessitating the development of robust hyperspectral analysis techniques to estimate apple-tree LNC. Additionally, the role of phenological information in improving LNC estimates across multiple growth stages remains uncertain. The present study thus proposes a novel hyperspectral data-processing framework (Boruta-Iteration-DNN) to optimize estimates of apple-tree LNC and investigates the potential of phenological information to improve LNC estimates of multi-fertility apple trees. The proposed framework incorporates the Boruta feature-selection method for identifying important spectral bands and assessing how they affect model performance through iterative feature analysis. A deep neural network (DNN) regression model is then developed for LNC estimation. In addition, a widely used Pearson-DNN method is used for comparison to demonstrate the superiority of the proposed approach. Furthermore, the potential of incorporating phenological information (Day After Anthesis, DAA) into the estimation models is examined to evaluate its ability to better estimate LNC. The results demonstrate that the proposed framework not only enhances the accuracy of apple-tree LNC estimates with respect to the full-spectrum model (without DAA: validation R2 improved from 0.69 to 0.75, NRMSE reduced from 11.92 % to 9.97 %; with DAA: validation R2 improved from 0.72 to 0.79, NRMSE reduced from 11.09 % to 9.06 %), but also reduces the complexity and dimensionality of the models by eliminating 96 % of the redundant spectral bands (2015 bands). Moreover, the inferior results of the Pearson-Iteration-DNN framework further underscore the effectiveness of the Boruta method in selecting relevant spectral bands. Additionally, including the DAA phenological information enhances the accuracy of all models (the total R2 of training and validation increased by 0.05 ∼ 0.12, and the total NRMSE reduced by 1.49 % ∼ 1.93 %). These findings highlight the ability of DNN models integrated with hyperspectral data and phenological information to accurately assess LNC in apple trees, thereby offering valuable guidance to orchard management for developing precise fertilization strategies.

RAAWC-UNet: an apple leaf and disease segmentation method based on residual attention and atrous spatial pyramid pooling improved UNet with weight compression loss.

Early detection of leaf diseases is necessary to control the spread of plant diseases, and one of the important steps is the segmentation of leaf and disease images. The uneven light and leaf overlap in complex situations make segmentation of leaves and diseases quite difficult. Moreover, the significant differences in ratios of leaf and disease pixels results in a challenge in identifying diseases. To solve the above issues, the residual attention mechanism combined with atrous spatial pyramid pooling and weight compression loss of UNet is proposed, which is named RAAWC-UNet. Firstly, weights compression loss is a method that introduces a modulation factor in front of the cross-entropy loss, aiming at solving the problem of the imbalance between foreground and background pixels. Secondly, the residual network and the convolutional block attention module are combined to form Res_CBAM. It can accurately localize pixels at the edge of the disease and alleviate the vanishing of gradient and semantic information from downsampling. Finally, in the last layer of downsampling, the atrous spatial pyramid pooling is used instead of two convolutions to solve the problem of insufficient spatial context information. The experimental results show that the proposed RAAWC-UNet increases the intersection over union in leaf and disease segmentation by 1.91% and 5.61%, and the pixel accuracy of disease by 4.65% compared with UNet. The effectiveness of the proposed method was further verified by the better results in comparison with deep learning methods with similar network architectures.

Analysis of flavonoid metabolism of compounds in succulent fruits and leaves of three different colors of Rosaceae

Red flesh apple (Malus pumila var. medzwetzkyana Dieck), purple leaf plum (Prunus cerasifera Ehrhar f), and purple leaf peach (Prunus persica ‘Atropurpurea’) are significant ornamental plants within the Rosaceae family. The coloration of their fruits and leaves is crucial in their appearance and nutritional quality. However, qualitative and quantitative studies on flavonoids in the succulent fruits and leaves of multicolored Rosaceae plants are lacking. To unveil the diversity and variety-specificity of flavonoids in these three varieties, we conducted a comparative analysis of flavonoid metabolic components using ultra-high-performance liquid phase mass spectrometry (UPLC-MS/MS). The results revealed the detection of 311 metabolites, including 47 flavonoids, 105 flavonols, 16 chalcones, 37 dihydroflavonoids, 8 dihydroflavonols, 30 anthocyanins, 14 flavonoid carbon glycosides, 23 flavanols, 8 isoflavones, 11 tannins, and 12 proanthocyanidins. Notably, although the purple plum and peach leaves exhibited distinct anthocyanin compounds, paeoniflorin and corythrin glycosides were common but displayed varying glycosylation levels. While the green purple leaf peach fruit (PEF) and red flesh apple leaf (AL) possessed the lowest anthocyanin content, they exhibited the highest total flavonoid content. Conversely, the red flesh apple fruit (AF) displayed the highest anthocyanin content and a diverse range of anthocyanin glycosylation modifications, indicating that anthocyanins predominantly influenced the fruit's color. Purple PLF, PLL, and PEL showcased varying concentrations of anthocyanins, suggesting that their colors result from the co-color interaction between specific types of anthocyanins and secondary metabolites, such as flavonols, flavonoids, and dihydroflavonoids. This study provides novel insights into the variations in tissue metabolites among Rosaceae plants with distinct fruit and leaf colors.

The Orchard Guard: Deep Learning powered apple leaf disease detection with MobileNetV2 model

The Orchard Guard: Deep Learning powered apple leaf disease detection with MobileNetV2 model

BLSENet: A Novel Lightweight Bilinear Convolutional Neural Network Based on Attention Mechanism and Feature Fusion Strategy for Apple Leaf Disease Classification

Accurate identification of apple leaf diseases is of great significance for improving apple yield. The lesion area of the apple leaf disease image is small and vulnerable to background interference, which easily leads to low recognition accuracy. To solve this problem, a lightweight bilinear convolutional neural network (CNN) model named BLSENet based on attention mechanism is designed. The model consists of two subnetworks, and each subnetwork is embedded with a Squeeze-and-Excitation (SE) module. By using the feature extraction ability of the two subnetworks and combining the bilinear feature CONCAT operation, the multiscale features of the image are obtained. Compared with the unimproved model LeNet-5 (84.63%), BLSENet has higher accuracy in the test set, which indicates that SE module and bilinear feature fusion have a positive effect on the performance of the model, and BLSENet has the ability to identify apple leaf diseases. The model has achieved the expected goal and can provide technical support for accurate identification and real-time monitoring of apple disease images.

Time-Course Responses of Apple Leaf Endophytes to the Infection of Gymnosporangium yamadae.

Apple rust, caused by Gymnosporangium yamadae, poses a significant challenge to apple production. Prior studies have underscored the pivotal role played by endophytic microbial communities, intimately linked with the host, in influencing plant diseases and their pathogenic outcomes. The objective of this study is to scrutinize alternations in endophytic microbial communities within apple leaves at different stages of apple rust using high-throughput sequencing technology. The findings revealed a discernible pattern characterized by an initial increase and subsequent decrease in the alpha diversity of microbial communities in diseased leaves. A microbial co-occurrence network analysis revealed that the complexity of the bacterial community in diseased leaves diminished initially and then rebounded during the progression of the disease. Additionally, employing the PICRUSt2 platform, this study provided preliminary insights into the functions of microbial communities at specific disease timepoints. During the spermogonial stage, endophytic bacteria particularly exhibited heightened activity in genetic information processing, metabolism, and environmental information processing pathways. Endophytic fungi also significantly enriched a large number of metabolic pathways during the spermogonial stage and aecial stage, exhibiting abnormally active life activities. These findings establish a foundation for comprehending the role of host endophytes in the interaction between pathogens and hosts. Furthermore, they offer valuable insights for the development and exploitation of plant endophytic resources, thereby contributing to enhanced strategies for managing apple rust.

Exploring the trend of recognizing apple leaf disease detection through machine learning: a comprehensive analysis using bibliometric techniques

This study’s foremost objectives were to scrutinize how unexpected weather affects agricultural output and to assess how well AI-based machine learning and deep leaning algorithms work for spotting apple leaf diseases. The researchers carried out a bibliometric study to obtain understanding of the current research trends, citation patterns, ownership and partnership arrangements, publishing patterns, and other parameters related to early identification of apple illnesses. Comprehensive interdisciplinary scientific maps are limited because syndrome recognition is not restricted to any solitary arena of research, despite the fact that there have been many studies on the identification of apple diseases. By employing a scientometric technique and 109 publications from the Scopus database published between 2011 and 2022, this study attempted to assess the condition of the research area and combine knowledge frameworks. To find important journals, authors, nations, articles, and topics, the study used the automated processes of VOSviewer and Biblioshiny software. Patterns and trends were discovered using citation counts, social network analysis, and citation and co-citation studies.

Retracted: Multidimensional Attention-Based CNN Model for Identifying Apple Leaf Disease

Retracted: Multidimensional Attention-Based CNN Model for Identifying Apple Leaf Disease

Morphogenesis Response of Leaf and Petiole Explant of Rootstock Apple (Malus sylvestris Mill.) to Auxin and Cytokinin

Rootstock apple is a type of plant used in grafting propagation. Research related to the response of morphogenesis in this plant has not been widely carried out. The aim of this research was to study the effect of auxin and cytokinin ratio on the morphogenesis response of leaf and petioles explants of rootstock apple and to study variations in the ratio of auxin and cytokinin with the best response. The study used a 2-factorial Completely Randomized Design (CRD) with variations in the types of explants and combinations of plants growth regulators (PGR) in the culture medium. There were 12 treatments with five repetitions. Explants in the form of young apple leaf (third leaf) and petiole rootstock that had been sterilized and cut with a size of 5x5 mm for the leaf and 5 mm long for the petiole were inoculated into MS medium with cytokinin in the form of BAP (1 ppm) combined with auxin, IBA (0, 0.1, 0.2, 0.3, 0.4, 0.5 ppm). Each culture bottle contains five explants. The results showed that the average leaf explant was capable of producing 34% callus. Meanwhile, petiole explants produced 4% callus. Callus began to form on leaf and petiole explants on the 26th and 42nd days after inoculation respectively. Most calluses have a compact texture with variations in green, yellowish white, and brownish green. The PGR combination with the highest percentage of callus formation in leaf explants was found in the BAP 1 ppm + IBA 0.4 ppm treatment, while the petiole was in the BAP 1 ppm + IBA 0.3 ppm treatment

A Dual-Branch Model Integrating CNN and Swin Transformer for Efficient Apple Leaf Disease Classification

Apples, as the fourth-largest globally produced fruit, play a crucial role in modern agriculture. However, accurately identifying apple diseases remains a significant challenge as failure in this regard leads to economic losses and poses threats to food safety. With the rapid development of artificial intelligence, advanced deep learning methods such as convolutional neural networks (CNNs) and Transformer-based technologies have made notable achievements in the agricultural field. In this study, we propose a dual-branch model named DBCoST, integrating CNN and Swin Transformer. CNNs focus on extracting local information, while Transformers are known for their ability to capture global information. The model aims to fully leverage the advantages of both in extracting local and global information. Additionally, we introduce the feature fusion module (FFM), which comprises a residual module and an enhanced Squeeze-and-Excitation (SE) attention mechanism, for more effective fusion and retention of both local and global information. In the natural environment, there are various sources of noise, such as the overlapping of apple branches and leaves, as well as the presence of fruits, which increase the complexity of accurately identifying diseases on apple leaves. This unique challenge provides a robust experimental foundation for validating the performance of our model. We comprehensively evaluate our model by conducting comparative experiments with other classification models under identical conditions. The experimental results demonstrate that our model outperforms other models across various metrics, including accuracy, recall, precision, and F1 score, achieving values of 97.32%, 97.33%, 97.40%, and 97.36%, respectively. Furthermore, detailed comparisons of our model’s accuracy across different diseases reveal accuracy rates exceeding 96% for each disease. In summary, our model performs better overall, achieving balanced accuracy across different apple leaf diseases.

L-Net: A lightweight convolutional neural network for devices with low computing power

Deep learning (DL) has demonstrated exceptional success across various domains, including computer vision, natural language processing, and speech recognition. However, the training and inference processes of DL models typically require substantial computational resources and storage space, presenting a significant challenge within the Internet of Things (IoT) domain. This study contributes theoretically to the field of lightweight DL by proposing L-Net, a lightweight convolutional neural network designed specifically for low-compute devices. The L-Net addresses challenges associated with channel interaction disparities and vanishing gradients. To further improve the network performance, we introduce the residual enhanced channel attention (or R-ECA) module, which combines a bypass mechanism derived from simplified residual learning with the attention mechanism's cross-channel interaction. Additionally, we replace the rectified linear unit function (or ReLU) with an exponential linear unit (or ELU) function to enhance the network's nonlinear expression capability and training speed. We conducted object recognition experiments and compared the accuracy and prediction stability of L-Net with well-known models, such as AlexNet, VGG11, SqueezeNet, ResNet, and MobileNet, to assess its efficacy. Using the CIFAR-10 dataset and our custom dataset of apple tree leaf diseases, our experimental results demonstrate that, with relatively smaller model parameters, L-Net performs exceptionally well in terms of mean Average Precision (mAP), achieving 0.906. Furthermore, when applied to our custom dataset, L-Net exhibits relatively consistent performance across various dataset splits under different ratios, outperforming the majority of models.

YOLOV5-CBAM-C3TR: an optimized model based on transformer module and attention mechanism for apple leaf disease detection.

Apple trees face various challenges during cultivation. Apple leaves, as the key part of the apple tree for photosynthesis, occupy most of the area of the tree. Diseases of the leaves can hinder the healthy growth of trees and cause huge economic losses to fruit growers. The prerequisite for precise control of apple leaf diseases is the timely and accurate detection of different diseases on apple leaves. Traditional methods relying on manual detection have problems such as limited accuracy and slow speed. In this study, both the attention mechanism and the module containing the transformer encoder were innovatively introduced into YOLOV5, resulting in YOLOV5-CBAM-C3TR for apple leaf disease detection. The datasets used in this experiment were uniformly RGB images. To better evaluate the effectiveness of YOLOV5-CBAM-C3TR, the model was compared with different target detection models such as SSD, YOLOV3, YOLOV4, and YOLOV5. The results showed that YOLOV5-CBAM-C3TR achieved mAP@0.5, precision, and recall of 73.4%, 70.9%, and 69.5% for three apple leaf diseases including Alternaria blotch, Grey spot, and Rust. Compared with the original model YOLOV5, the mAP 0.5increased by 8.25% with a small change in the number of parameters. In addition, YOLOV5-CBAM-C3TR can achieve an average accuracy of 92.4% in detecting 208 randomly selected apple leaf disease samples. Notably, YOLOV5-CBAM-C3TR achieved 93.1% and 89.6% accuracy in detecting two very similar diseases including Alternaria Blotch and Grey Spot, respectively. The YOLOV5-CBAM-C3TR model proposed in this paper has been applied to the detection of apple leaf diseases for the first time, and also showed strong recognition ability in identifying similar diseases, which is expected to promote the further development of disease detection technology.

Seasonal trends in leaf and fruit nutrient concentrations of apple (Malus x domestica Borkh.) grafted on different clonal rootstocks and their effect on fruit quality

Effective nutrient management necessitates monitoring of nutrient requirement, extent and timing of nutrient supply during the fruit development. To achieve a good nutrient balance and its impact on fruit quality, study of nutrient dynamics is of utmost importance as it provides an intriguing stratagem to control tree nutrient status. The research aimed at investigating seasonal patterns of nutrient concentrations in leaves and fruits of apple (Malus × domestica Borkh.) cv Silver Spur on various clonal rootstocks (MM106, M4 and M9) during the year 2019–2020 and their impact on fruit quality. For this purpose collection of leaf and fruit samples commenced from one month after full bloom (May 15) until harvesting (September 18) at an interval of twenty one days and samples were evaluated for both macro and micro-nutrients {Nitrogen (N), Phosphorus (P), Potassium (K), Calcium (Ca), Magnesium (Mg) and Boron (B)} and quality characteristics. In general, concentration of N, P, K and B declined continuously in both leaf and fruit toward the end of fruit maturation, whereas leaf Ca and Mg augmented along the cycle, however fruit Ca and Mg increased up to first fruit development stage and then decreased again signifying nutrient dilution along growing season and its effect on fruit quality. With respect to rootstocks, semi-vigorous rootstock MM106 recorded higher values for leaf and fruit nutrient concentrations as compared to dwarf M9 and semi-dwarf M4 rootstocks specifying a significant effect of rootstocks on nutrient absorption.

Deep learning architectures for diagnosing the severity of apple frog-eye leaf spot disease in complex backgrounds.

In precision agriculture, accurately diagnosing apple frog-eye leaf spot disease is critical for effective disease management. Traditional methods, predominantly relying on labor-intensive and subjective visual evaluations, are often inefficient and unreliable. To tackle these challenges in complex orchard environments, we develop a specialized deep learning architecture. This architecture consists of a two-stage multi-network model. The first stage features an enhanced Pyramid Scene Parsing Network (L-DPNet) with deformable convolutions for improved apple leaf segmentation. The second stage utilizes an improved U-Net (D-UNet), optimized with bilinear upsampling and batch normalization, for precise disease spot segmentation. Our model sets new benchmarks in performance, achieving a mean Intersection over Union (mIoU) of 91.27% for segmentation of both apple leaves and disease spots, and a mean Pixel Accuracy (mPA) of 94.32%. It also excels in classifying disease severity across five levels, achieving an overall precision of 94.81%. This approach represents a significant advancement in automated disease quantification, enhancing disease management in precision agriculture through data-driven decision-making.

Exploration of a multifunctional biocontrol agent Streptomyces sp. JCK-8055 for the management of apple fire blight.

Apple fire blight, caused by the bacterium Erwinia amylovora, is a devastating disease of apple and pear trees. Biological control methods have attracted much attention from researchers to manage plant diseases as they are eco-friendly and viable alternatives to synthetic pesticides. Herein, we isolated Streptomyces sp. JCK-8055 from the root of pepper and investigated its mechanisms of action against E. amylovora. Streptomyces sp. JCK-8055 produced aureothricin and thiolutin, which antagonistically affect E. amylovora. JCK-8055 and its two active metabolites have a broad-spectrum in vitro activity against various phytopathogenic bacteria and fungi. They also effectively suppressed tomato bacterial wilt and apple fire blight in in vivo experiments. Interestingly, JCK-8055 colonizes roots as a tomato seed coating and induces apple leaf shedding at the abscission zone, ultimately halting the growth of pathogenic bacteria. Additionally, JCK-8055 can produce the plant growth regulation hormone indole-3-acetic acid (IAA) and hydrolytic enzymes, including protease, gelatinase, and cellulase. JCK-8055 treatment also triggered the expression of salicylate (SA) and jasmonate (JA) signaling pathway marker genes, such as PR1, PR2, and PR3. Overall, our findings demonstrate that Streptomyces sp. JCK-8055 can control a wide range of plant diseases, particularly apple fire blight, through a combination of mechanisms such as antibiosis and induced resistance, highlighting its excellent potential as a biocontrol agent. KEY POINTS: • JCK-8055 produces the systemic antimicrobial metabolites, aureothricin, and thiolutin. • JCK-8055 treatment upregulates PR gene expression in apple plants against E. amylovora. • JCK-8055 controls plant diseases with antibiotics and induced resistance.

Hemijski sastav infuzije divlje jabuke (Malus sylvestris (L.) Mill.)

Wild apples leaves are under-utilized biomaterial, containing various bioactive compounds. The aim of this work was to determine the components and mineral composition of the wild apple leaf infusion. The leaves originated from Svrljig, Nišava District, Southern Serbia. The screening of bioactive components in the infusion was performed by the UHPLC-DAD-MS/MS method, while its mineral content was analyzed by the ICP-OES method. The results indicated the presence of quercetin, isoquercetin, hyperoside, rutin, citric and quinic acids, protocatechuic acid, 3-O-p-coumaroylquinic acid, chlorogenic acid, procyanidin B2, phloretin and phloretin-pentosyl-hexoside, while the most abundant elements were Ca, K, Mg and Na. Therefore, wild apple leaves may be utilized for isolation of polyphenolic antioxidants and minerals and production of extracts with antioxidant activity.