Enhancing Tea Leaf Disease Identification with Lightweight MobileNetV2

Diseases in tea trees can result in significant losses in both the quality and quantity of tea production.Regular monitoring can help to prevent the occurrence of large-scale diseases in tea plantations.However,existingmethods face challenges such as a high number of parameters and low recognition accuracy,which hinders their application in tea plantation monitoring equipment.This paper presents a lightweight I-MobileNetV2 model for identifying diseases in tea leaves,to address these challenges.The proposed method first embeds a Coordinate Attention(CA)module into the originalMobileNetV2 network,enabling the model to locate disease regions accurately.Secondly,a Multi-branch Parallel Convolution(MPC)module is employed to extract disease features across multiple scales,improving themodel’s adaptability to different disease scales.Finally,the AutoML for Model Compression(AMC)is used to compress themodel and reduce computational complexity.Experimental results indicate that our proposed algorithm attains an average accuracy of 96.12%on our self-built tea leaf disease dataset,surpassing the original MobileNetV2 by 1.91%.Furthermore,the number of model parameters have been reduced by 40%,making itmore suitable for practical application in tea plantation environments.

Detection and Identification of Phytoplasma of Cleome rutidosperma in Areca Palm Yellow Leaf Disease Field

[Objectives]The paper was to detect and identify the phytoplasma of Cleome rutidosperma in areca palm yellow leaf disease(YLD)field in Wenchang City,Hainan Province,China.[Methods]The nested PCR technique was employed to amplify the phytoplasma 16S rDNA of C.rutidosperma samples,followed by sequence analysis.Concurrently,this study examined C.rutidosperma in YLD field,collecting symptomatic leaves for phytoplasma detection.[Results]The 16S rDNA sequence of the C.rutidosperma witches'-broom phytoplasma was found to be identical to that of the HNWC5 strain associated with areca palm yellows phytoplasma,leading to the identification of this phytoplasma as belonging to the 16SrII-A subgroup.Field investigations revealed a higher incidence of C.rutidosperma in areca palm fields,with symptoms of leaf yellows observed in six of these fields.Quantitative PCR(qPCR)analysis confirmed the presence of phytoplasma infection in these instances.[Conclusions]Through the analysis of geographical distribution,sequence alignment,and field occurrence data,a significant correlation has been identified between witches'broom disease and YLD.It is proposed that the former may act as an intermediate host for the areca palm yellows phytoplasma.

A Novel Agricultural Data Sharing Mode Based on Rice Disease Identification

In this paper,a variety of classical convolutional neural networks are trained on two different datasets using transfer learning method.We demonstrated that the training dataset has a significant impact on the training results,in addition to the optimization achieved through the model structure.However,the lack of open-source agricultural data,combined with the absence of a comprehensive open-source data sharing platform,remains a substantial obstacle.This issue is closely related to the difficulty and high cost of obtaining high-quality agricultural data,the low level of education of most employees,underdeveloped distributed training systems and unsecured data security.To address these challenges,this paper proposes a novel idea of constructing an agricultural data sharing platform based on a federated learning(FL)framework,aiming to overcome the deficiency of high-quality data in agricultural field training.

Image Generation of Tomato Leaf Disease Identification Based on Small-ACGAN

Plant diseases have become a challenging threat in the agricultural field.Various learning approaches for plant disease detection and classification have been adopted to detect and diagnose these diseases early.However,deep learning entails extensive data for training,and it may be challenging to collect plant datasets.Even though plant datasets can be collected,they may be uneven in quantity.As a result,the problem of classification model overfitting arises.This study targets this issue and proposes an auxiliary classifier GAN(small-ACGAN)model based on a small number of datasets to extend the available data.First,after comparing various attention mechanisms,this paper chose to add the lightweight Coordinate Attention(CA)to the generator module of Auxiliary Classifier GANs(ACGAN)to improve the image quality.Then,a gradient penalty mechanism was added to the loss function to improve the training stability of the model.Experiments show that the proposed method can best improve the recognition accuracy of the classifier with the doubled dataset.On AlexNet,the accuracy was increased by 11.2%.In addition,small-ACGAN outperformed the other three GANs used in the experiment.Moreover,the experimental accuracy,precision,recall,and F1 scores of the five convolutional neural network(CNN)classifiers on the enhanced dataset improved by an average of 3.74%,3.48%,3.74%,and 3.80%compared to the original dataset.Furthermore,the accuracy of MobileNetV3 reached 97.9%,which fully demonstrated the feasibility of this approach.The general experimental results indicate that the method proposed in this paper provides a new dataset expansion method for effectively improving the identification accuracy and can play an essential role in expanding the dataset of the sparse number of plant diseases.

Lightweight Method for Plant Disease Identification Using Deep Learning

In the deep learning approach for identifying plant diseases,the high complexity of the network model,the large number of parameters,and great computational effort make it challenging to deploy the model on terminal devices with limited computational resources.In this study,a lightweight method for plant diseases identification that is an improved version of the ShuffleNetV2 model is proposed.In the proposed model,the depthwise convolution in the basic module of ShuffleNetV2 is replaced with mixed depthwise convolution to capture crop pest images with different resolutions;the efficient channel attention module is added into the ShuffleNetV2 model network structure to enhance the channel features;and the ReLU activation function is replaced with the ReLU6 activation function to prevent the gen-eration of large gradients.Experiments are conducted on the public dataset PlantVillage.The results show that the proposed model achieves an accuracy of 99.43%,which is an improvement of 0.6 percentage points compared to the ShuffleNetV2 model.Compared to lightweight network models,such as MobileNetV2,MobileNetV3,EfficientNet,and EfficientNetV2,and classical convolutional neural network models,such as ResNet34,ResNet50,and ResNet101,the proposed model has fewer parameters and higher recognition accuracy,which provides guidance for deploying crop pest identification methods on resource-constrained devices,including mobile terminals.

Investigation and Identification of Fungal Diseases of Kiwifruit in Qiandongnan Prefecture

[Objectives]The paper was to investigate and identify the fungal diseases of wild and red heart kiwifruit in Qiandongnan Prefecture.[Methods]The pathogenic fungi were isolated from diseased leaves and fruits of wild and red heart kiwifruit by tissue separation method.DNA sequencing was carried out by using the sequence analysis of ribosomal r DNA-ITS region,and molecular evolutionary trees were built by using MEGA 4.0 software.Finally,the pathogenic fungi were classified and identified by combining morphological observation.[Results]The main fungal diseases were anthracnose caused by Colletotrichum gloeosporioides on wild kiwifruit,fruit anthracnose caused by C.acutatum on red heart kiwifruit,leaf soft rot caused by Fusarium incarnatum on red heart kiwifruit,and brown spot caused by Alternaria alternata on red heart kiwifruit.[Conclusions]The study may provide some theoretical basis for the control of kiwifruit diseases in Qiandongnan Prefecture.

Establishment and Application of Identification Methods for Resistance to Sugarcane White Leaf Disease(SCWL)

[Objectives]This study was conducted to establish simple, efficient, stable, standardized and practical identification methods for sugarcane resistance to white leaf disease(SCWL), and promote the breeding for sugarcane resistance to SCWL. [Methods]The identification technology of sugarcane resistance to SCWL was systematically studied and explored from the aspects of sugarcane material treatment and planting, inoculation liquid preparation, inoculation method, disease investigation, grading standard formulation, etc., and two sets of simple, efficient, stable, standardized and practical accurate identification methods for sugarcane resistance to SCWL were created for the first time, namely, the seed cane coating inoculation method and the stem-cutting inoculation method at the growth stage. The seed cane coating inoculation method includes the steps of directly screening SCWL phytoplasma, extracting juice from cane and adding 10 times of sterile water to prepare an inoculation liquid, spraying seed cane on plastic film to keep moisture, planting the inoculated materials in barrels in an insect-proof greenhouse for cultivation, investigating the incidence rate 30 d after inoculation, and evaluating the disease resistance according to the 1-5 level standard. The method of stem-cutting inoculation includes the steps of directly screening sugarcane stems carrying SCWL phytoplasma and adding 10 times of sterile water to prepare an inoculation liquid, cultivating the identification materials in an insect-proof greenhouse, dropping 100 μl of the inoculation liquid into each root incision with a pipette gun at the age of 6 months, investigating the incidence rate 20 d after planting, and evaluating the disease resistance according to the 1-5 level standard. [Results] The two methods are similar to the natural transmission method. After inoculation, SCML occurred significantly, with high sensitivity and good reproducibility. The results of resistance identification were consistent with those of natural disease in the field. Through the two inoculation methods and field natural disease investigation, the resistance of 10 main cultivars to SCML was identified, which was true and reliable. [Conclusions] This study can provide standard varieties for identification of SCML resistance in the future.

Research on Identification Method of Apple Diseases in Southern Xinjiang Based on Deep Learning and Its System Implementation

Apple disease samples were collected from the southern Xinjiang and annotated to design a convolutional neural network model based on deep learning.The accuracy and robustness of the model was improved through training and optimization algorithms,and a complete apple disease identification system was developed with the model as the core,and evaluated for its performance in terms of accuracy,recall rate and speed.This study provides a reliable AI-based apple disease diagnosis solution for the apple planting industry in the southern Xinjiang,hoping to help farmers better manage and protect crop health.

Isolation and Identification of Antifungal Bacterial Strain KL-1 against Plant Wilt Disease

[Objective] The antifungal bacteria of plant wilt disease was screened and identified to provide foundation for the study on bio-control preparation of plant wilt disease.[Method] Confrontation culture method was adopted to screen the bio-control bacteria with good antifungal effect against plant wilt disease,Biolog bacteria automatic identification system and 16S rDNA sequence analysis method were selected to identify its taxonomic status,the biological safety of the strain towards cotton and mice was also determined.[Result] 12 bacterial strains were isolated from rhizosphere of cotton.Among those strains,5 isolates showed antifungal activity against F.decemcellulare Brick,F.oxysporum f.sp.Diathi,F.oxysporum f.sp.vasinfectum.The antifungal effect of KL-1 strain against three target strains of pathogen reached 69.09%,80.78% and 78.89% respectively.Identification results of Biolog bacteria automatic identification system and 16S rDNA sequence analysis method showed that KL-1strain was Bacillus amyloliquefaciens;primary determination results of biological safety also showed that the strain KL-1 was safe and non-toxic towards cotton and mice.[Conclusion] KL-1strain of B.amyloliquefaciens had antifungal effect against several pathogens of plant wilt diseases,which was safe and non-toxic towards cotton and mice,being the bio-control strain with research and development potential.

Identification of a Tobacco Disease on Flat Tray System

[ Objective] The pathogen of a new disease found in floating seedlings of tobacco was isolated and identified to provide the basis for the control of the disease. [ Method] The pathogenicity of the isolated strain was determined according to Koch＇s rules, and the pathogen was identified through Biolog system, 16S rDNA sequence analysis and physiological and biochemical methods. [ Result] Through Koch＇s test, the isolated 3 -3 strain was verified to be the pathogen causing floating seedling disease, which was consistent with the characteristics of Pectobacterium carotovo- rum subsp. Carotovorum through Biolog determination and the other physiological and biochemical methods. 16S rDNA sequence analysis showed that 3 -3 strain had the highest similarity with P. carotovorum subsp. Carotovorum strains Kun28213 （accession number GU936996）, reaching 99. 9%. [Conclusion] Base on the identification results of several methods, the pathogen causing floating seedling disease was P. carotovorum subsp. Carotovorum, and the disease was first reported in China. According to the English name of the disease, the disease was called as tobacco blackleg disease.

Investigation of Main Fungal Diseases of Pepper (Piper nigrum L.) in Hainan Province and Their Pathogen Identification

[ Objective ] The paper was to confirm the current major diseases of pepper in Hainan Province and their corresponding pathogens. [ Method ] The pep- per gardens in 13 main cultivation regions of pepper in Hainan Province were systematically investigated, and the pathogens of the obtained specimens were isolated and identified. [ Result] Seven fungal diseases in pepper were totally investigated, including blast, anthracnose, blight, slow wilt, root rot, basal rot and scleretin- ia disease. Currently, the diseases with serious damage on pepper include blast, anthracnese, blight and slow wilt. [ Conclusion] The result provides the theoreti- cal basis for the integrated control of diseases in pepper, and is also benefit for scientific research workers to master the latest dynamic of diseases.

Deep Transfer Learning Models for Mobile-Based Ocular Disorder Identification on Retinal Images

Mobile technology is developing significantly.Mobile phone technologies have been integrated into the healthcare industry to help medical practitioners.Typically,computer vision models focus on image detection and classification issues.MobileNetV2 is a computer vision model that performs well on mobile devices,but it requires cloud services to process biometric image information and provide predictions to users.This leads to increased latency.Processing biometrics image datasets on mobile devices will make the prediction faster,but mobiles are resource-restricted devices in terms of storage,power,and computational speed.Hence,a model that is small in size,efficient,and has good prediction quality for biometrics image classification problems is required.Quantizing pre-trained CNN(PCNN)MobileNetV2 architecture combined with a Support Vector Machine(SVM)compacts the model representation and reduces the computational cost and memory requirement.This proposed novel approach combines quantized pre-trained CNN(PCNN)MobileNetV2 architecture with a Support Vector Machine(SVM)to represent models efficiently with low computational cost and memory.Our contributions include evaluating three CNN models for ocular disease identification in transfer learning and deep feature plus SVM approaches,showing the superiority of deep features from MobileNetV2 and SVM classification models,comparing traditional methods,exploring six ocular diseases and normal classification with 20,111 images postdata augmentation,and reducing the number of trainable models.The model is trained on ocular disorder retinal fundus image datasets according to the severity of six age-related macular degeneration(AMD),one of the most common eye illnesses,Cataract,Diabetes,Glaucoma,Hypertension,andMyopia with one class Normal.From the experiment outcomes,it is observed that the suggested MobileNetV2-SVM model size is compressed.The testing accuracy for MobileNetV2-SVM,InceptionV3,and MobileNetV2 is 90.11%,86.88%,and 89.76%respectively while MobileNetV2-SVM,InceptionV3,and MobileNetV2 accuracy are observed to be 92.59%,83.38%,and 90.16%,respectively.The proposed novel technique can be used to classify all biometric medical image datasets on mobile devices.

Identification of Pathogen of Gastrodia elata White Silk Disease

[ Objective] The aim was to identify the pathogen of Gastrodia elata white silk disease and explore prevention and treatment of the disease.[ Method] Infected G. elata, collected from low-mountain regions in Yichang city of Hubei Province, were isolated and purified. Then, Koch＇s Postulate was adopted to inoculate the pathogen. Finally, the identification was carried out according to its biological characteristics. [ Result] The pathogen of white silk disease belongs to weak hyperparasite. Because of its poor direct invasion ability, the pathogen usually invades into host through wound infection. According to its hyphae morphology and pathogenicity, the pathogen of G. elata White Silk Disease has been identified as Sc/erotium ro/fsii Sacc., which belongs to deuteromycotina, hyphomycetes, agonomycetales, fungi of sclerotium genus. [ Conclusion ] The study will provide references for the control of G. elata White Silk Disease caused by S. rolfsii.

Screening and Identification of Antagonistic Bacterium ZG-10 against Spot Blotch Disease in Pakchoi

[ Objective] The purpose was to screen bacterium with antagonistic effect against pathogen of spot blotch disease in pakchoi in vegetable field. [Method] More than 200 strains of bacteria which could produce spore were isolated from soil in different places. Through screening and rescreening, the bacteria with higher antibacterial activity were conducted observation about thallus shapes and colony characters, a series of physiological and biochemical tests were performed. [Result] Rescreening results indicated that the strains including ZG-10, ZG-19, ZG-59, ZG-72 and ZG-31 had significant antibacterial activity, which had very high research value and good application prospect for biocontrol on spot blotch disease in pakchoi; the strain ZG-10 was identified to be Bacillus subtilis. [ Conclusion] The strain ZG-10 had biocontrol potential and good development prospect. This research laid certain basis for subsequent research and strated a new way for the application of antagonistic strain and proteinum polypeptide in agdculture.

Identification of Crop Diseases Based on Improved Genetic Algorithm and Extreme Learning Machine

As an indispensable task in crop protection,the detection of crop diseases directly impacts the income of farmers.To address the problems of low crop-disease identification precision and detection abilities,a new method of detection is proposed based on improved genetic algorithm and extreme learning machine.Taking five different typical diseases with common crops as the objects,this method first preprocesses the images of crops and selects the optimal features for fusion.Then,it builds a model of crop disease identification for extreme learning machine,introduces the hill-climbing algorithm to improve the traditional genetic algorithm,optimizes the initial weights and thresholds of the machine,and acquires the approximately optimal solution.And finally,a data set of crop diseases is used for verification,demonstrating that,compared with several other common machine learning methods,this method can effectively improve the crop-disease identification precision and detection abilities and provide a basis for the identification of other crop diseases.

Pathogen Isolation and Identification of a Serious Leaf Disease of Red Sandalwood (Pterocarpus santalinus) Seedlings

In the introduction and propagation of red sandalwood （Pterocarpus santalinus）, a serious leaf disease of its seedlings in winter and spring seasons was found, but the name of the disease and its pathogen species have not been reported. The pathogen isolated from infected leaves of 18-month-old seedlings was identi- fied as Colletotrichum gloeosporioides by morphological characteristics of colony and conidium, and analysis results of rDNA-intemal transcribed spacer sequence （ITS） of the strain. Pathogenicity test results further confirmed that C. gloeosporioides was the pathogen responsible for the infected leaves symptoms of red sandal- wood. However, the disease belongs to an atypical anthraenose. Control of the leaf diseases of red sandalwood seedlings was discussed.

Isolation and Identification of a Specific cDNA Mapping to the Bam HI-I2 and -LFragments within the Inverted Repeats ofUnique Long Re-gion (IRL) in the Genom e ofMarek′s Disease Herpesvirus (MDV) Oncogenic Strain Beijing-1

Objective\ To understand the transcription of BamHI L DNA fragment from genome of strong virulent GA strain of Marek′s disease herpesvirus (MDV) in lymphoblastoid tumor tissue induced by oncogenic strain Beijing 1 (a specific local strain in China) of MDV. Methods\ Two oligonucleotide primers were synthesized according to the reported sequence of \%meq\% gene an ideal oncogenic candidate and our previously determined sequence of BamHI L fragment of Marek′s disease herpesvirus (MDV), respectively. Reverse transcriptase PCR(RT PCR) assay was performed by using these primers and the mRNA as a template which was isolated from visceral lymphoblastoid tumors obtained from chickens artificially infected with strain Beijing 1 of oncogenic MDV. Southern blot molecular hybridization was further carried out to detect the product of RT PCR with digoxigenin labeled nucleotide probe from BamHI I2 and L fragment in the gene library of MDV strain GA, respectively. Results\ Two probes could simultaneously hybridize this cDNA amplified by RT PCR with a length of about 730 bp. Conclusion\ It is suggested that \%meq\% transcription could extend from the right hand end of BamHI I2 to the adjacent BamHI L, and the BamHI L region was likely to be transcribed in MDV induced lymphoblastoid tumors.

Types of Maize Virus Diseases and Progress in Virus Identification Techniques in China

There are a total of more than 40 reported maize viral diseases worldwide. Five of them have reportedly occurred in China. They are maize rough dwarf disease, maize dwarf mosaic disease, maize streak dwarf disease, maize crimson leaf disease, maize wallaby ear disease and corn lethal necrosis disease. This paper reviewed their occurrence and distribution as well as virus identification techniques in order to provide a basis for virus identification and diagnosis in corn production.

Identification and Control of HLB Disease in Citrus grandis

HLB disease has been endangering citrus production,and it is an important task in Citrus grandis production to identify and control HLB disease. Diaphorina citri and Trioza erytreae are the main vectors for the spreading of HLB disease. Scientific and proper release of predatory natural enemies such as ladybugs,combined with chemical control can effectively control psyllid. For suspected HLB disease strains,a simple " HLB disease detection reagent" can be used for detection. This method is simple,cheap and accurate,and it is an economical and feasible identification method for ordinary growers.

Identification and Control of Radish Alternaria Leaf Spot, Black Rot, Soft Rot, Virus Disease and Hollowness

The damage characteristics and occurrence regularity of radish alternaria leaf spot, black rot, soft rot, virus disease and hollowness are introduced in the paper, and the corresponding control method is proposed from two aspects of agricultural control and chemical control.