This document summarizes recent research on detecting and classifying crop leaf diseases using machine learning and deep learning techniques. It reviews papers that use image processing, machine learning and deep learning to analyze visual symptoms in leaf images. Convolutional neural network models have achieved notable accuracies above 90% in identifying diseases. The document also discusses limitations and opportunities for improving real-time automated disease detection systems.
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as TXT, PDF, TXT or read online on Scribd
0 ratings0% found this document useful (0 votes)
56 views9 pages
Book
This document summarizes recent research on detecting and classifying crop leaf diseases using machine learning and deep learning techniques. It reviews papers that use image processing, machine learning and deep learning to analyze visual symptoms in leaf images. Convolutional neural network models have achieved notable accuracies above 90% in identifying diseases. The document also discusses limitations and opportunities for improving real-time automated disease detection systems.
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as TXT, PDF, TXT or read online on Scribd
You are on page 1/ 9
Institute of Advanced Engineering and Science
International Journal of Electrical and Computer Engineering (IJECE)Vol. 12, No. 2,
April2022, pp. 2079~2086ISSN: 2088-8708, DOI: 10.11591/ijece.v12i2.pp2079- 20862079Journal homepage: http://ijece.iaescore.comCrop leaf disease detection and classification using machine learning and deep learning algorithms by visual symptoms: a reviewPallepati Vasavi1,2, Arumugam Punitha1, T.Venkat Narayana Rao21Department of Computer Science and Engineering, Annamalai University, Chidambaram, Tamilnadu,India2Department of Computer Science and Engineering, Sreenidhi Institute of Science and Technology, Hyderabad, Telangana, IndiaArticle InfoABSTRACTArticle history:Received Apr21, 2021Revised Jul18, 2021Accepted Aug 10, 2021A quick and precise crop leaf disease detection is important to increasing agricultural yield in a sustainable manner. We present a comprehensive overview of recent research in the field of crop leaf disease prediction using image processing (IP), machine learning(ML)and deep learning (DL)techniques in this paper. Using these techniques, crop leaf disease prediction made it possible to get notable accuracies. This article presents a survey of research papers that presented the various methodologies, analyzes them in terms of the dataset, number of images, number of classes, algorithms used, convolutional neural networks (CNN)models employed, and overall performance achieved. Then, suggestions are prepared on the most appropriate algorithms to deploy in standard,mobile/embedded systems, drones, robots and unmanned aerial vehicles (UAV). We discussed the performance measures used and listed some of the limitations and future works that requires to be focus on, to extend real time automated crop leaf disease detection system.Keywords:Convolutional neural networksDeep learningImage processingMachine learningPlant disease detectionVisual symptomsThis is an open access article under the CC BY-SAlicense.Corresponding Author:Pallepati VasaviDepartment of Computer Science and Engineering,Sreenidhi Institute of Science and Technology Hyderabad, Telangana, IndiaEmail: [email protected] is an agrarian economy, with Arable land accounting for more than 54% of the total land area. India ranks among the world's top producers of rice, wheat, cotton, fruits and vegetables, and dairy products in terms of volume. The demand for agricultural products is growing at an unprecedented rate as our population grows. Good nutrition make sure that human body gets all of the nutrients, vitamins, and minerals it needs to work optimally. Maintaining essential vitamins and minerals is also essential for good health. Preventing and monitoring crop diseases plays crucial role for providing nutrias food. Since they can damage crops, decreasing food supply and chain while also raising food prices. Plant pests and diseases can also reduce the palatability of foods, causing populations to alter their conventional food preferences. In 1970, a disease in soya bean crops, “Sudden Death Syndrome” rapidly increased across the United States(US)and eventually reaching on the whole agricultural areas of the US. So, quick and precise crop leaf disease recognition is dangerous to increasing agricultural yield in a sustainable manner. The motto of Food and Agriculture Organization of United Nations(FAO)is“let be bread”.Crop leaf diseases vary in shape, size, and color. Certain diseases might have the identical color, but dissimilar shapes; while some have dissimilar colors but identical shapes. The model can be developed by capturing the diseased leaves and recognize the patterns about the disease is helpful to get free of crop loss ISSN:2088-8708Int J Elec & Comp Eng, Vol. 12, No. 2, April2022: 2079-20862080due to disease spread or increase[1]. In this approach, the images are often sent to a core crop leaf disease system for analysis; the system can recognize. The system generates information about the crop leaf disease.In order to explore the use of Image processing for classifying citrus leaf diseases, a research[2]on 2006, was conducted. For this analysis, four distinct citrus leaf disease groups were used, namely greasy spot, melanose, healthy, and scab. For feature extraction and classification the proposed algorithms based on image-processing techniques were developed. The process of extraction of features used the technique of color co-occurrence, which usesan image's colorand texture to achieve specific characteristics that reflect the image. On all data models using intensity features, SAS discriminant analysis, hue and saturation features, hyperspectral image (HSI)features presented the results above 81percent, and above 95.8 percent respectively.In 2007 Tellaeche et al. [3]proposed two mechanisms: segmentation of images and decision making. To divide cells from the image as low-level parts, image segmentation incorporates simple image processing techniques. 2 area-basedattributes computing the relationships between crop rows and weeds define each part. A hybrid supervisory methodology decides, from these properties, whether or not a cell must be sprayed. The decision is depends on the merger of two well-known classifiers(SVMand FM) under the Bayesian system.To recognize cucumber crop leaf disease based on IP and SVM was introduced in 2008[1]. To reduce noise from the obtained cucumber disease leaves color images, thevector median filter was initially used. The color picture of the cucumber disease spot on the leaf was derived from the texture, shape and color characteristics. The system used SVM and neural networks classifiers. Shape feature gives more accuracy than the texture and color feature gives faster results. The results showed that SVM performance is better than neural networks. Linear kernel of SVM gives better results than polynomial, radial basic, and sigmoid functions.In 2014,Gavhale et al[4]developed a system that contains four-part image preprocessing model involving (red, green, blue) RGB to different colorconversion, image enhancement techniques; segmenting the region of interest (ROI)using K-means clustering for algebraic use to assess the defect and acute areas of crop leaves, extraction and classification of features. Using statistical GLCMand color function by means of mean values, texture feature extraction. Finally, the classification obtained using SVM (polynomial andRBF).The proposed model in 2015, Mokhtar etal.[5]uses theGLCM to detect, whether safe or contaminated, and to classify tomato leaf status. For the classification process, the SVM algorithm with popular kernel functions is used. Datasets of 800 healthy and diseased tomato leaves in total. The N-fold cross-validation technique was used to test the accuracy of the presented method with 99.83 percent accuracy of classification using linear kernel functionality.Mohanty et al.[6]proposed a system in 2016 to recognize plant leaf diseases using DL. They built a deep CNNto classify 14 crop varieties and 26 diseases by a dataset of 54,306 images of PlantVillage dataset. They compared various CNN architectures on transfer learning and scratch training. In the case of the colored version of the dataset, the models work better. The limitation is that the classification of single leaves, facing up, in a homogeneous context, is currently limited.Rançonet al.[7]compared SIFTencoding and DL feature extractors which are already pre trained, for the recognition of Esca disease in Vineyards. 91 percent overall accuracy was obtained using deep extracted features from the ImageNet database trained MobileNet network, exhibiting the efficacy of transfer learning methods without the need to build afeature extractor of ad-hoc specialized features. The next part was aimed at the identification of diseases (using bounding boxes) inside complete images of the plant. The deep learning core network has been incorporated into a "one-step" detection network (RetinaNet) for this reason, enabling us to execute recognition queries in near real time (approximately 6 frames per second on GPU).This paper [8]proposesa deep neural network-based real time detection system for pests and diseases of Cole. A bounding box generator first determines to provide bounding boxes of size, location, and class by training the input image with a region-based neural network. Then, for verification, the presenting bounding boxes from each class are fed into the CNN filter bank. The problem of false positives generated by bounding box generators and class inequalities in data sets with incomplete data can be solved by proposed method.Goncharov etal.[9]to provide the solution for the problem of tiny image databases, the deep siamese convolutional network was created. The identification of the 3 diseases namely Esca, Black rot, andChlorosis disease on grape leaves had an accuracy of over 90%.Panigrahi et al.[10]the research focused on traditional machine learning techniques for the recognition of maize crop diseases, such as NB, DT, KNN, SVM, and RF. In order to choose the most apt model with the highest precision for plant disease prediction, the aforementioned classification techniques are analyzed and compared. The RF algorithm provides the best Int J Elec & Comp EngISSN:2088-8708Crop leaf disease detection and classification using machine learning and deep... (Pallepati Vasavi)2081results, with 79.23% accuracy.Sagar et al.[11]compared five different architectures: ResNet50,VGG16,InceptionV3, Inception, ResNet, and DenseNet169.They discovered that the best result on the test set was ResNet50 with 94% accuracy, it uses skip connections using a residual layer archive. They computed performance by precision, accuracy, recall and F1 score.2.RESEARCH METHOD To detect crop leaf diseases and classification by visual symptoms, there are 5 steps existed in the crop leaf disease detection and classification model architecture: image acquisition, image preprocessing, image segmentation, feature extraction, classification. Table1 represents various acronyms used in the crop leaf disease detection. Table 2 interprets the model of crop leaf disease detection and classification.Table 1. AcronymsAcronymAbbreviationAcronymAbbreviationMLMachine learningKNNK nearest neighborsDLDeep learningRFRandom forestIPImage processingSURFSpeed up robust featuresCNNConvolutional neural networksLRLogistic regressionUAVUnmanned aerial vehiclesBPNNBack propagation neural networksFAOFood and agriculture organizationANNArtificial neural networksSVMSupport vector machineR-FCNRegion-based fully convolutional networkFMFuzzy K meansR-CNNRegionbased convolutional neural networksGLCMGary level co-occurrence matrixLBPLocal binary patternsRBFRadial basis functionSGDStochastic gradient descentHSVHue, saturation, and valueHOGHistogram of an oriented gradientSIFT encodingScale invariant feature transformGPUGraphical processing unitNBNaïve BayesDTDecision treeTable 2. List of the methods used for the detection of crop leaf diseasesImage acquisitionImage PreprocessingImage SegmentationFeature ExtractionClassificationCapturing the images from drones, smart mobiles phones, digital cameras and UAVsImage augmentationK-means, Principal component analysis ClusteringTexture, shape and color are the featuresMachine LearningCollecting the images from public datasetsImage resizing, rotations, Flipping, shift, shear, zoomThresholdingRGB feature extractionPNN, SVM, ANN, RBF, KNN, BPNN, NN, DT, RF, NBPlantVillageImage AnnotationsColor segmentationcolor co- occurrenceDeep LearningImage database of plant disease symptoms (PDDB)Image EnhancementLearning based segmentationGLCM texture extractionCNN, Optimized CNNBugwood image database systemRemoving noiseEdge detectionSIFTLSTMWheat Disease Database 2017SmoothingModel based segmentationSURFTransfer LearningIPM ImagesHistogram equalizationforeground/backgroundHOGVGG19, GoogLeNet, AlexNet, ResNet50, Inception_V3, MobileNetKaggleMedian filteringOtsu thresholdingNASNetUCI repositoryColor transformationsSobel edge detectionSqueezeNetcontrast Image Enhancementsemantic segmentationDeep Siamese Neural Networksperspective, affine image transformationscontours-based segmentationEnsemble ModelsClippingF- RCNN,SSD,R-FCN2.1.Image acquisitionThis is the first step of crop leaf disease detection and classification. The purpose of this stage is to collect and prepare images dataset that will be used in the further process. This is done by capturing the ISSN:2088-8708Int J Elec & Comp Eng, Vol. 12, No. 2, April2022: 2079- 20862082images from mobile phone cameras, digital cameras, drones and UAV either on real time (site) or in controlled conditions.2.2. Image preprocessingImage preprocessing is very important to obtain the better results. To remove the noise colortransformations were used.To reducethe size of the image acquired by digital cameras resizing techniques were used. It also helps to reduce memory size. The frequently used image preprocessingtechniques in this literature includes cropping the leaves from the acquired images, color transformations, rescaling, background removal, image enhancement, flipping, rotating, Shear,and image smoothing.2.3. Image segmentationImage segmentation plays anessential role in crop leaf disease detection and classification. It splits the image into various parts or zones. It explores the image data to extract helpful information for feature extraction. Image segmentation can be done in 2 ways, one is based on similarities and the other one is based on discontinuities.2.4. Feature extractionExtracting the features of the substances of an image is called as feature extraction. The most common features found in plant disease detection and classifications are shape, color, and texture. The crop diseases may differ in appearances of the image due to multiple classes. The crop leaf disease system can easily recognize the diseases from the shape of the crop leaf image. The second feature is color is an important. It distinguishes the crop leaf diseases from each other. The last feature, texture portrays thevarious patterns of the color are spotted in the crop leaf images. The common texture features are energy, entropy, contrast, correlation, sum of squares, sum entropy, cluster shade, clusterprominence, homogeneity. 2.5. ClassificationTwo types of classification methods were used to classify crop leaf diseases:ML andDL. The important dissimilarity between traditional machine learning and deep learning methods is by means of feature extraction. In traditional ML, the features are not computed automatically whereas in DL the feature extraction automatically takes place and it is considered as learning weights. So, in DL the system itself learns the needed features by providing sufficient data.The most common machine learning algorithms used for classification of plant diseases are KNN, SVM, DT, RF, BPNN, NN, NB and ensemble learning. The frequently used deep learning algorithms present in the literature were CNN, CNN models which were Pre trained on ImageNet and used transfer learning.Chowdhuryet al. [12]proposed a plant disease detection and classification system, it uses transfer learning and deep feature extraction in. The authors were compared the obtained results of VGG16, GoogLeNet, ResNet50 CNN architectures with deep feature extraction by SVM and KNN. Experiment results shown that classification with SVM and ResNet50 given best results (98%) than the remaining combinations. The authors also compared the results of traditional machine learning algorithms i.e.SVM and KNN, SVM shown better accuracy (80.6%) than KNN(71.8%) but it is lesser than the proposed.3.RESULTS AND DISCUSSIONIt is observed from the Table 3 is the number of images used for the detection of crop leaf diseases by machine learning techniques are very less compared to deep learning techniques but generated remarkable accuracies.”represents that the information is not mentioned in the paper. By using modified CNNs, optimized deep learning models and transfer learning models gives better results than the basic CNNs. Modified DL techniques gives better performance than traditional ML techniques. Modified CNN i.e. Multi-channelmodel gives highest accuracy[13]i.e., 99.5% in DL and SVM with linear kernel gives accuracy of 99% in ML techniques. Figure 1(a)and Figure 1(b)represents the performance of the crop disease prediction system by using DL, ML techniques. The most used performance measures mentioned in the survey are accuracy, precision, K-fold cross validation technique for example k=10, recall, F1-score, sensitivity, specificity, dice similarity coefficient (DSC), minimum square error (MSE), and structural similarity index measurement (SSIM), categorical cross-entropy, Matthew’s correlation coefficient (MCC) and leave one out cross validation scheme. Table 3 interprets the details of the various researches on crop leaf disease prediction and classification. Int J Elec & Comp EngISSN:2088-8708Crop leaf disease detection and classification using machine learning and deep... (Pallepati Vasavi)2083(a)(b)Figure 1. The performance of the crop disease prediction system (a) performance of the DL techniquesand(b) performance of the ML techniquesTable 3. Details of the surveyed papers for the detection and classification of Crop leaf diseasesYearReference NoCropNumber of imagesNumber of ClassesAlgorithmAccuracy2018[14]____2SVM,KNN,RF,NB,LR582019[15]Multiple800SVM-Multi class652018[16]Papaya160__RF,SVM,LR,LDA,NAÏVE BAYES,KNN,CART702020[17]Multiple5430527Faster R-CNN with Inception Resnet v270.532020[10]Maize34234Naïve Baye's77.462020[12]Grape622865Faster R- CNN,Inception-v1-ResNet-v2,SE blocks and RPN81.12019[18]Mulberry__3CNN822017[19]Cotton9007SVM(regression)832020[20]papaya100 003ResNet852018[21]Tomato14007CNN86.92018[22]Paddy__Alex Net872018[23]Multiple__38CNN88.62019[9]Grape1304Deep Siamese convolution network902019[24]Potato24652Faster R- CNN902020[25]tomato4,6713MobileNet902020[26]Multiple148,77538Inception v3 transferred to target domain SVM90.62020[27]Tomato1000010CNN91.22016[28]Cucumber3004global-local SVD(single value decomposition)--SVM classifier91.632019[29]Maize1004CNN92.852016[30]Alfalfa8994SVM94.72016[31]alfalfa89 94SVM,RF,KNN94.742013[32]Multiple50030SVM952019[33]Wheat81784ResNet50962018[34]Cass ava7603GMLVQ972020[35]Multiple54,30538VGG1697.82020[36]Multiple54,30538VGG1697.8202 0[37]Maize68665Optimized VGG1697.92020[38]Maize68665Ensemble model of two pre- trained CNN97.92020[38]20,00019CNN982019[39]Multiple120,000ResidualNet982020[11]Multiple54, 30538Pre trained ResNet50,98.22020[40]Maize11523CNN model with optimized trainable parameters98.42019[35]Tomato1792910CNN98.62020[41]__400CNN(ReLu,Tanh,Softsign,linea r,sigmoid)98.82015[5]Tomato____SVM(linear kernel)992017[42]Tomato602NN992019[43]Banana18,00018Faster R-CNN/SSD+ResNet50992019[44]Apple11924Alex Net, Google Net and DenseNet201 Layer99.22018[45]Multiple8500058CNN99.52019[13]Multiple5400038Multichannel CNN99.52018[46]Potato3003SVM__2018[47]Lentil3002LBP,HBBP (Brightness Bi-Histogram Equalization)__2021[48]Multiple54,30538CNN__2016[49]Brinjal__4ANN__2018[50]Soyabean 162079CNN__5060708090100NNRandom ForestSVM(RBF)KNNGMLVQSVMSVM(Regression)Naïve BayesSVM(Gaussian kernel)BPNNSVM(linear kernel)Accuracy5060708090100Deep siamese...Small CNNMobileNet V2ResNet50CNNFaster R-...GoogLeNetMultichannel CNNVGG16NASNetInception_V3Accuracy ISSN:2088-8708Int J Elec & Comp Eng, Vol. 12, No. 2, April2022: 2079- 208620844.LIMITATIONS AND FUTURE WORKS The most common limitations of crop leaf disease prediction system based on visual symptoms are Lack of sufficient datasets mentioned in the papers since the PlantVillage is the only public dataset that is generated under controlled conditions. Some of the authors developed their own dataset but they are not giving the access to others to compare the results and PlantVillagedataset could not provide the images for the commercial crops like chiliwith abundant number of diseases. The second most common limitation is the method of developing the dataset by image acquisition and image preprocessing. The common problems are Un-even illumination, Clutter field background and real cultivation condition and other parts of the plant images were not used to detect the plant diseases. Thirdly, apply the knowledge of ensemble algorithms, tuning of hyper parameters and diversity of pooling operations [19]. Fourthly, the prediction system needs enormous resources if the prediction was based on deep learning methodologies [25]. So, there is a significance to develop squeeze models to run the application in mobile phones, drones, UAVs and robots.The frequent future works were, to develop real time enormous images and classes of plant diseases. Crop disease dataset can be integrated to incorporate location, weather and soil data of the diseased plant to examine crop and yield monitoring in support of smart agriculture. The crop disease prediction system can be enhanced for detecting the plant diseases in large scale horticultural fields.5.CONCLUSIONThe system for identifying the crop leaf diseases can be developed in 5 steps. These are including Image acquisition;image pre-processing, image segmentation, feature extraction and classification. In this survey we analyzed various methodologies versus accuracies, datasets, crops, requirements of number of images and classes, also analyzed performance measures, limitations and future works. The conclusion of the study enhances the importance of integrating computer vision, machine learning, deep learning to the automated devices like UAVs, smart mobiles in the era of agriculture. More investigations, datasets have to be developed to detect the diseases at the instant time even if the yield is large scale and containing multiple diseases also, with the squeezable resources.REFERENCES [1]T. Youwen, L. Tianlai, and N. Yan, “The recognition of cucumber disease based on image processing and support vector machine,” in 2008 Congress on Image and Signal Processing, 2008, pp. 262–267, doi: 10.1109/CISP.2008.29.[2]R. Pydipati, T. F. Burks, and W. S. Lee, “Identification of citrus disease using color texture features and discriminant analysis,” Computers and Electronics in Agriculture, vol. 52, no. 1–2, pp. 49–59, Jun. 2006, doi: 10.1016/j.compag.2006.01.004.[3]A. Tellaeche, X. P. Burgos-Srtizzu, G. Pajares, and A. Ribeiro, “On combining support vector machines and fuzzy kmeans in vision based precision agriculture,” World Academy of Science, Engineering and Technology, International Journal of Computer, Electrical, Automation, Control and Information Engineering, vol. 28, no. 4, pp. 33–38, 2007.[4]K. R. Gavhale, U. Gawande, and K. O. Hajari, “Unhealthy region of citrus leaf detection using image processing techniques,” in International Conference for Convergence for Technology-2014, Apr. 2014, pp. 1–6, doi: 10.1109/I2CT.2014.7092035.[5]U. Mokhtar et al., “SVM-Based Detection of Tomato Leaves Diseases,” in Frontiers in Plant Science, vol. 7, 2015, pp. 641–652.[6]S. P. Mohanty, D. P. Hughes, and M. Salathé, “Using deep learning for image-based plant disease detection,” Frontiers in Plant Science, vol. 7, pp. 1–10, Sep. 2016, doi: 10.3389/fpls.2016.01419.[7]F. Rançon, L. Bombrun, B. Keresztes, and C. Germain, “Comparison of SIFT encoded and deep learning features for the classification and detection of esca disease in bordeaux vineyards,” Remote Sensing, vol. 11, no. 1, pp. 1–26, Dec. 2018, doi: 10.3390/rs11010001.[8]Z. Libo, H. Tian, G. Chunyun, and M. Elhoseny, “Real-time detection ofcole diseases and insect pests in wireless sensor networks,” Journal of Intelligent & Fuzzy Systems, vol. 37, no. 3, pp. 3513–3524, Oct. 2019, doi: 10.3233/JIFS-179155.[9]P. Goncharov, G. Ososkov, A. Nechaevskiy, A. Uzhinskiy, and I. Nestsiarenia, “Disease Detection on the Plant Leaves by Deep Learning,” in Advances in Neural Computation, Machine Learning, and Cognitive Research II. NEUROINFORMATICS 2018. Studies in Computational Intelligence, 2019, pp. 151–159.[10]K. P. Panigrahi, H. Das, A. K. Sahoo,and S. C. Moharana, “Maize leaf disease detection and classification using machine learning algorithms,” in Progress in Computing, Analytics and Networking, 2020, pp. 659–669.[11]A. Sagar and D. Jacob, “On using transfer learning for plant disease detection,” bioRXiv, 2020, doi: 10.13140/RG.2.2.12224.15360/1.[12]X. Xie, Y. Ma, B. Liu, J. He, S. Li, and H. Wang, “A deep-learning-based real-time detector for grape leaf diseases using improved convolutional neural networks,” Frontiers in Plant Science, vol. 11, no. 751, pp. 1–14, Jun. 2020, doi: 10.3389/fpls.2020.00751.[13]G. Anthonys and N. Wickramarachchi, “An image recognition system for crop disease identification of paddy fields in Sri Lanka,” in 2009 International Conference on Industrial and Information Systems (ICIIS), Dec. 2009, pp. 403–407, doi: 10.1109/ICIINFS.2009.5429828. [14]M. H. Jumat, M. S. Nazmudeen, and A. T. Wan, “Smart farm prototype for plant disease detection, diagnosis & treatment using IoT device in a greenhouse,” in 7th Brunei International Conference on Engineering and Technology 2018 (BICET 2018), 2018, pp. 1–4, doi: 10.1049/cp.2018.1545.[15]S. Poornima, S. Kavitha, S. Mohanavalli, and N. Sripriya, “Detection and classification of diseases in plants using image processingand machine learning techniques,” 2019, doi: 10.1063/1.5097529.[16]S. Ramesh et al., “Plant disease detection using machine learning,” in 2018 International Conference on Design Innovations for Int J Elec & Comp EngISSN:2088-8708Crop leaf disease detection and classification using machine learning and deep... (Pallepati Vasavi)20853Cs Compute Communicate Control (ICDI3C), Apr. 2018, pp. 41–45, doi: 10.1109/ICDI3C.2018.00017.[17]D. Singh, N. Jain, P. Jain, P. Kayal, S. Kumawat, and N. Batra, “PlantDoc: a dataset for visual plant disease detection,” in Proceedings of the 7th ACM IKDD CoDS and 25th COMAD, Jan. 2020, pp. 249–253, doi: 10.1145/3371158.3371196.[18]D. D. Hema, S. Dey, Krishabh, and A. Saha, “Mulberry leaf disease detection using deep learning,” International Journal of Engineering and Advanced Technology (IJEAT), vol. 9, no. 1, pp. 3366–3371, 2019, doi: 10.35940/ijeat.A1521.109119.[19]A. A. Sarangdhar and V. R. Pawar, “Machine learning regression technique for cotton leaf disease detection and controlling using IoT,” in 2017 International conference of Electronics, Communication and Aerospace Technology (ICECA), Apr. 2017, pp.449–454, doi: 10.1109/ICECA.2017.8212855.[20]R. K. Veeraballi, M. S. Nagugari, C. S. R. Annavarapu, and E. V. Gownipuram, “Deep learning based approach for classificationand detection of papaya leaf diseases,” in Intelligent Systems Design and Applications. ISDA 2018 2018. Advances in Intelligent Systems and Computing, 2020, pp. 291–302.[21]A. Hidayatuloh, M. Nursalman, and E. Nugraha, “Identification of tomato plant diseases by leaf image using squeezenet model,”in 2018 International Conference on Information Technology Systems and Innovation (ICITSI), Oct. 2018, pp. 199–204, doi: 10.1109/ICITSI.2018.8696087.[22]A. A. Alfarisy, Q. Chen, and M. Guo, “Deep learning based classification for paddy pests & diseases recognition,” in Proceedings of 2018 International Conference on Mathematics and Artificial Intelligence, Apr. 2018, pp. 21–25, doi: 10.1145/3208788.3208795.[23]R. Gandhi, S. Nimbalkar, N. Yelamanchili, and S. Ponkshe, “Plant disease detection using CNNs and GANs as an augmentative approach,” in 2018 IEEE International Conference on Innovative Research and Development (ICIRD), May 2018, pp. 1–5, doi: 10.1109/ICIRD.2018.8376321.[24]D. Oppenheim, G. Shani, O. Erlich, and L. Tsror, “Using deep learning for image-based potato tuber disease detection,” Phytopathology®, vol. 109, no. 6, pp. 1083–1087, Jun. 2019, doi: 10.1094/PHYTO-08-18-0288-R.[25]S. Z. M. Zaki, M. Asyraf Zulkifley, M. Mohd Stofa, N. A. M. Kamari, and N. Ayuni Mohamed, “Classification of tomato leaf diseases using MobileNet v2,” IAES International Journal of Artificial Intelligence (IJ-AI), vol. 9, no. 2, pp. 290–296, Jun. 2020, doi: 10.11591/ijai.v9.i2.pp290-296. [26]D. Argüeso et al., “Few-Shot Learning approach for plant disease classification using images taken in the field,” Computers and Electronics in Agriculture, vol. 175, Aug. 2020, doi: 10.1016/j.compag.2020.105542.[27]M. Agarwal, A. Singh, S. Arjaria, A. Sinha, and S. Gupta, “ToLeD: tomato leaf disease detection using convolution neural network,” Procedia Computer Science, vol. 167, pp. 293– 301, 2020, doi: 10.1016/j.procs.2020.03.225.[28]S. Zhang and Z. Wang, “Cucumber disease recognition based on global-local singular value decomposition,” Neurocomputing, vol. 205, pp. 341–348, Sep. 2016, doi: 10.1016/j.neucom.2016.04.034.[29]M. Sibiya and M. Sumbwanyambe, “A computational procedure for the recognition and classification of maize leaf diseases out of healthy leaves using convolutional neural networks,” AgriEngineering, vol. 1, no. 1, pp. 119–131, Mar. 2019, doi: 10.3390/agriengineering1010009.[30]J. Francis, Anto Sahaya Dhas D, and Anoop B K, “Identification of leaf diseases in pepper plants using soft computing techniques,” in 2016 Conference on Emerging Devices and Smart Systems (ICEDSS), Mar. 2016, pp. 168–173, doi: 10.1109/ICEDSS.2016.7587787.[31]F. Qin, D. Liu, B. Sun, L. Ruan, Z. Ma, and H. Wang, “Identification of alfalfa leaf diseases using image recognition technology,” PLOS ONE, vol. 11, no. 12, Dec. 2016, doi: 10.1371/journal.pone.0168274.[32]S. Arivazhagan, R. N. Shebiah, S. Ananthi, and S. V. Varthini, “Detection of unhealthy region of plant leaves and classification of plant leaf diseasesusing texture features,” Agricultural Engineering International: CIGR Journal, vol. 15, no. 1, pp. 211–217, 2013.[33]A. Picon, A. Alvarez-Gila, M. Seitz, A. Ortiz-Barredo, J. Echazarra, and A. Johannes, “Deep convolutional neural networks for mobile capture device-based crop disease classification in the wild,” Computers and Electronics in Agriculture, vol. 161, pp. 280–290, Jun. 2019, doi: 10.1016/j.compag.2018.04.002.[34]G. Owomugisha, F. Melchert, E. Mwebaze, J. A. Quinn, and M. Biehl, “Machine learning for diagnosis of disease in plants using spectral data,” 2018.[35]P. Sharma, Y. P. S. Berwal, and W. Ghai, “Performance analysis of deep learning CNN models for disease detection in plants using image segmentation,” Information Processing in Agriculture, vol. 7, no. 4, pp. 566–574, Dec. 2020, doi: 10.1016/j.inpa.2019.11.001.[36]F. Mohameth, C. Bingcai, and K. A. Sada, “Plant disease detection with deep learning and feature extraction using PlantVillage,” Journal of Computer and Communications, vol.08, no. 06, pp. 10–22, 2020, doi: 10.4236/jcc.2020.86002.[37]A. Darwish, D. Ezzat, and A. E. Hassanien, “An optimized model based on convolutional neural networks and orthogonal learning particle swarm optimization algorithm for plant diseases diagnosis,” Swarm and Evolutionary Computation, vol. 52, Feb. 2020, doi: 10.1016/j.swevo.2019.100616. [38]P. Sharma, P. Hans, and S. C. Gupta, “Classification of plant leaf diseases using machine learning and image preprocessing techniques,” in 2020 10th International Conference on Cloud Computing, Data Science & Engineering (Confluence), Jan. 2020, pp. 480–484, doi: 10.1109/Confluence47617.2020.9057889. [39]K. R., H. M., S. Anand, P. Mathikshara, A. Johnson, and M. R., “Attention embedded residual CNN for diseasedetection in tomato leaves,” Applied Soft Computing, vol. 86, Jan. 2020, doi: 10.1016/j.asoc.2019.105933.[40]C. R. Rahman et al., “Identification and recognition of rice diseases and pests using convolutional neural networks,” Biosystems Engineering, vol. 194, pp. 112–120, Jun. 2020, doi: 10.1016/j.biosystemseng.2020.03.020.[41]R. B. S., T. A. Shriram, J. S. Raju, M. Hari, B. Santhi, and G. R. Brindha, “Farmer-friendly mobile application for automated leaf disease detection of real-time augmented data set using convolution neural networks,” Journal of Computer Science, vol. 16, no. 2, pp. 158–166, Feb. 2020, doi: 10.3844/jcssp.2020.158.166.[42]J. F. Molina, R. Gil, C. Bojaca, F. Gomez, and H. Franco, “Automatic detection of early blight infection on tomato crops using a color based classification strategy,” in 2014 XIX Symposium on Image, Signal Processing and Artificial Vision, Sep. 2014, pp. 1–5, doi: 10.1109/STSIVA.2014.7010166.[43]M. G. Selvaraj et al., “AI-powered banana diseases and pest detection,” Plant Methods, vol. 15, no. 1, Dec. 2019, doi: 10.1186/s13007-019-0475-z.[44]M. Turkoglu, D. Hanbay, and A. Sengur, “Multi-model LSTM-based convolutional neural networks for detection of apple diseases and pests,” Journal of Ambient Intelligence and Humanized Computing, Nov. 2019, doi: 10.1007/s12652-019-01591-w.[45]K. P. Ferentinos, “Deep learning models for plant disease detection and diagnosis,” Computers and Electronics in Agriculture, vol. 145, pp. 311–318, Feb. 2018, doi: 10.1016/j.compag.2018.01.009. [46]M. Islam, Anh Dinh, K. Wahid, and P. Bhowmik, “Detection of potato diseases using image segmentation and multiclasssupport ISSN:2088-8708Int J Elec & Comp Eng, Vol. 12, No. 2, April2022: 2079- 20862086vector machine,” in 2017 IEEE 30th Canadian Conference on Electrical and Computer Engineering (CCECE), Apr. 2017, pp. 1–4, doi: 10.1109/CCECE.2017.7946594. [47]K. Singh, S. Kumar, and P. Kaur, “Automatic detection of rust disease of Lentil by machine learning system using microscopic images,” International Journal of Electrical and Computer Engineering (IJECE), vol. 9, no. 1, pp. 660–666, Feb. 2019, doi: 10.11591/ijece.v9i1.pp660-666.[48]D. Varshney, B. Babukhanwala, J. Khan, D. Saxena, and Ashutoshkumarsingh, “Machine learning techniques for plant disease detection,” 2021.[49]R. Anand, S. Veni, and J. Aravinth, “An application of image processing techniques for detection of diseases on brinjal leaves using k-means clustering method,” in 2016 International Conference on Recent Trends in Information Technology (ICRTIT), Apr. 2016, pp. 1–6, doi: 10.1109/ICRTIT.2016.7569531.[50]S. Ghosal, D. Blystone, A. K. Singh, B. Ganapathysubramanian, A. Singh, and S. Sarkar, “An explainable deep machine vision framework for plant stress phenotyping,” Proceedings of the National Academy of Sciences, vol. 115, no. 18, pp. 4613–4618, May 2018, doi: 10.1073/pnas.1716999115.BIOGRAPHIES OF AUTHORS Pallepati Vasavireceived Graduation (B. Tech) in 2012 from Computer Science and Engineering and Post Graduation (M. Tech) in 2014 from Software Engineering at JNTUH, Hyderabad, and Telangana. She is currently working as Assistant Professor in Sreenidhi Institute of Science and Technology, Hyderabad, Telangana. She is also pursuing Ph. D from Annamalai University, Chidambaram, and Tamilnadu. Her research interests include Machine Learning, Deep Learning and Image Processing. Email: [email protected] Punitha, Associate Professor, Computer Science and Engineering, Annamalai University, Chidambaram, and Tamilnadu. She Pursued her Ph. D from Annamalai University in Computer Science and Engineering, Annamalai University, Chidambaram, and Tamilnadu.Her research interests include Image and video Processing. Email: [email protected] Narayana Rao Professor of C.S.E, SNIST, is B.E in Computer Science and Engineering from Nagpur University; he has completed M.Tech (CSE) from SIT, JNTUH, Campus, and Hyderabad. He holds a Ph.D from JNTUK, Kakinada. He has completed M.B.A with Systems and Marketing as specialization.He has industry experience as a software engineer in HCL Technologies Ltd, Bangalore and has worked in high-end projects such as “Singapore Provident Fund” and “WISPA”. His total Experience spans 28 years. His research interests include Image Processing, Information Security, Data Mining and Software Enigeering. Email:[email protected]. Related papers thumb Cassava Leaf Disease Identification and Detection Using Deep Learning Approach Provided by: Agora University Editing House: Journals | Publisher: 'Agora University of Oradea' | Year: 2022 by Anitha J., Saranya N. thumb Techniques of deep learning and image processing in plant leaf disease detection: a review Provided by: Institute of Advanced Engineering and Science | Publisher: 'Institute of Advanced Engineering and Science' | Year: 2023 by Kini Anita S., Reddy Prema K. V., Pai Smitha N. thumb Image based Plant leaf disease detection using Deep learning Provided by: Sri Shakthi SIET Journals | Publisher: 'IOR Press' | Year: 2021 by S Poornam, A Francis Saviour Devaraj thumb Machine learning for plant disease detection: an investigative comparison between support vector machine and deep learning Provided by: Universiti Teknologi Malaysia Institutional Repository | Publisher: 'Institute of Advanced Engineering and Science' | Year: 2020 by Muhammad Abdu Aliyu, Mohd. Mokji Musa, Ullah Sheikh Usman thumb Corn leaf image classification based on machine learning techniques for accurate leaf disease detection Provided by: Institute of Advanced Engineering and Science | Publisher: 'Institute of Advanced Engineering and Science' | Year: 2022 by Noola Daneshwari Ashok, Rangapura Basavaraju Dayanand thumb Machine learning for detection and prediction of crop diseases and pests: A comprehensive survey Provided by: Repositório Institucional do ISCTE-IUL | Publisher: 'MDPI AG' | Year: 2022 by Domingues T., Brandão T., Ferreira J.