Classification Of Rice Plant Diseases Based on Leaf Images Using the Multi Class Support Vector Machine (M-SVM) Method

− In agriculture, rice plays an important role in the Indonesian economy. Rice produces rice, one of the most widely consumed staple food sources in Indonesia. Many factors can cause rice production failure, one of which is leaf pests and diseases. Therefore, early identification and management of plant diseases is an important step in an effort to increase crop yields and ensure food safety. One way to detect rice leaf images early is to perform an image classification process and create a web-based application. The method that has the ability in image processing is deep learning technique with convolutional neural network (CNN) method. The Convolutional Neural Network (CNN) method works to perform and predict diseases in plants by using image categorization or object images. This research aims to apply the web application of image classification of rice plant diseases to the Amazon Web Service (AWS) by identifying and classifying various types of rice leaf diseases using the CNN algorithm, so that farmers can detect rice plant diseases quickly and accurately through image analysis. This application was created using Convolutional Neural Network (CNN) methodology and Software Development Life Cycle (SDLC). The result of this study is that researchers created a web application for the classification of rice plant diseases through leaf images which are divided into 4 categories, namely Healthy, Leaf Blight, Brown Leaf Blight and Hispa, which is made a classification model using CNN with an accuracy value of 0. 8608, then using the streamlit framework to build a website, and utilizing AWS services in the form of Amazon Elastic Compute Cloud (Amazon EC2) as a hosting service, Amazon Simple Storage Service (Amazon S3) as a service for storing rice plant disease classification models and for storing web files, and Amazon Identity and Access Management Role (Amazon IAM) as a service to create a role that gives permission to connect between AWS S3 and AWS EC2. Testing the disease classification model in rice plants implemented on the web in EC2 shows quite good results with an accuracy of 78.5%. This can affect the model's ability to recognize specific disease patterns

The detection of diseases in rice plants is an essential step in ensuring healthy crop growth and maximizing yields. A real-time and accurate plant disease detection technique can assist in the development of mitigation strategies to ensure food security on a large scale and economical rice crop protection. An accurate classification of rice plant diseases using DL and computer vision could create a foundation to achieve a site-specific application of agrochemicals. Image investigation tools are efficient for the early diagnosis of plant diseases and the continuous monitoring of plant health status. This article presents an Enhanced Sea Horse Optimization with Deep Learning-based Multimodal Fusion for Rice Plant Disease Detection and Classification (ESHODL-MFRPDC) technique. The proposed technique employed a DL-based fusion process with a hyperparameter tuning strategy to achieve an improved rice plant disease detection performance. The ESHODL-MFRPDC approach used Bilateral Filtering (BF)-based noise removal and contrast enhancement as a preprocessing step. Furthermore, Mayfly Optimization (MFO) with a Multi-Level Thresholding (MLT) based segmentation process was used to recognize the diseased portions in the leaf image. A fusion of three DL models was used for feature extraction, namely Residual Network (ResNet50), Xception, and NASNet. The Quasi-Recurrent Neural Network (QRNN) was used for the recognition of rice plant diseases, and its hyperparameters were set using the ESHO method. The performance of the ESHODL-MFRPDC method was validated using the rice leaf disease dataset from the UCI database. An extensive comparison study demonstrated the promising performance of the proposed method over others.

Abstract. Early and accurate diagnosis of plant diseases is a vital step in the crop protection system. In traditional practices, identification is performed either by visual observation or by testing in laboratory. The visual observation requires expertise and it may vary subject to an individual which may lead to an error while the laboratory test is time consuming and may not be able to provide the results in time. To overcome these issues, image based machine learning approach to detect and classify plant diseases has been presented in literature. We have focused specifically on rice plant (Oryza sativa) disease in this paper. The images of the diseased symptoms in leaves and stems have been captured from the rice field. We have collected a total of 619 rice plant diseased images from the real field condition belong to four classes:(a) Rice Blast (RB), (b) Bacterial Leaf Blight (BLB), (c) Sheat Blight (SB) and (d) Healthy Leave (HL). We have used a pre-trained deep convolutional neural network(CNN) as a feature extractor and Support Vector Machine (SVM) as a classifier. We have obtained encouraging results. The early identification of rice diseases by this approach could be used as a preventive measure well as an early warning system. Further, it could be extended to develop a rice plant disease identification system on real agriculture field.

In a previous paper, the author (2001) proved the convergence of a commonly used decomposition method for support vector machines (SVMs). However, there is no theoretical justification about its stopping criterion, which is based on the gap of the violation of the optimality condition. It is essential to have the gap asymptotically approach zero, so we are sure that existing implementations stop in a finite number of iterations after reaching a specified tolerance. Here, we prove this result and illustrate it by two extensions: /spl nu/-SVM and a multiclass SVM by Crammer and Singer (2001). A further result shows that, in final iterations of the decomposition method, only a particular set of variables are still being modified. This supports the use of the shrinking and caching techniques in some existing implementations. Finally, we prove the asymptotic convergence of a decomposition method for this multiclass SVM. Discussions on the difference between this convergence proof and the one in another paper by Lin are also included.

Rice plant diseases are among the most critical problems that are being faced by the farmers. The diseases affect the quality and quantity of the crop, which impacts the economy of countries like India, where agriculture is the primary occupation. Therefore, an early and accurate identification of plant disease is crucial to get the maximum yield from the crop. Traditionally, identifying plant diseases by observing or testing them in the laboratory is time-consuming. Many researchers have worked on image-based machine learning (ML) approach for detection and classification of plant diseases. This paper presented an ML-based Support Vector Machine (SVM) kernel technique for detecting diseases in rice plants. Classification is done using SVM with different hyperparameters (SVM kernels and regularization parameter) for the early and critical assessment of rice plants. This paper concludes that the SVM model trained with the optimized parameters obtained the highest accuracy of 0.996, which is better than the previous techniques and reveals the novelty of the work.

In agriculture, technology can provide benefits to farmers. However, at present there are still very few farmers who use technology, especially computerization in their agricultural processes, such as the identification of diseases in rice plants, there are still many rice farmers who cannot recognize and distinguish the types of diseases in rice plants. Research on the identification of bacterial leaf blight and brown spots on rice plants has carried out before, but the accuracy rate is only 70%. This research developed a system to identify bacterial leaf blight and brown spot in rice plants through leaf images with an image processing approach. Image of affected rice leaves is segmented first using K-Means Clustering, then the texture features are extracted using the Gray Level Co-Occurrence Matrix (GLCM) with features extracted in the form of energy, contrast, correlation, homogeneity and shape pattern characteristics using metric and eccentricity features, then identified using Euclidean Distance. The training data used 40 images for each disease and 12 images for each disease. The test results show that the system has a better level of accuracy than previous studies that reached 100% with a Mean Squared Error (MSE) value of 0.007282214.

Identification of diseases from the images of a plant is one of the interesting research areas in the agriculture field, for which machine learning concepts of computer field can be applied. This article presents a prototype system for detection and classification of rice diseases based on the imag es of infected rice plants. This prototype system is developed after detailed experimental analysis of various techniques used in image processing operations. We consider three rice plant diseases namely Bacterial leaf blight, Brown spot, and Leaf smut. We capture images of infected rice plants using a digital camera from a rice field. We empirically evaluate four techniques of background removal and three techniques of segmentation. To enable accurate extraction of features, we propose centroid feeding based K-means clustering for segmentation of disease portion from a leaf image. We enhance the output of K-means clustering by removing green pixels in the disease portion. We extract various features under three categories: color, shape, and texture. We use Support Vector Machine (SVM) for multi-class classification. We achieve 93.33% accuracy on training dataset and 73.33% accuracy on the test dataset. We also perform 5 and 10-fold cross-validations, for which we achieve 83.80% and 88.57% accuracy, respectively.

This research aims to classify diseases in rice plants using the K-Nearest Neighbor (K-NN) algorithm based on Hue Saturation Value (HSV) color feature extraction and Gray Level Co-Occurrence Matrix (GLCM) texture. The main problem faced is how to identify the type of disease in rice plants automatically using digital images. Diseases such as Blight, Tungro, and Crackle often attack rice plants and require an accurate early detection system. Lack of understanding in recognizing disease symptoms manually often leads to errors in handling. For this reason, this research develops an image processing-based classification system that can detect diseases such as Blight, Tungro, and Crackle. The method used in this research is image processing which includes RGB to HSV color space conversion, texture feature extraction using GLCM, and classification using K-NN algorithm. The dataset consists of 240 images, divided into training data and testing data, namely 192 training data and 48 testing data. Tests were conducted by calculating accuracy at various values of the K parameter, namely K = 1, K = 3, and K = 5, to determine the effectiveness of the model in classifying plant diseases. The purpose of this study was to evaluate the accuracy of the system in identifying rice diseases and test the combination of HSV and GLCM features in improving classification performance. The results showed that using HSV and GLCM features together resulted in the highest accuracy at K=3 with an accuracy value of 75%. The system is expected to assist farmers in detecting plant diseases quickly and effectively, thus minimizing production losses and supporting agricultural sustainability

For horizontal circular pipes under uniform wall heat flux boundary condition and three different inlet configurations (re-entrant, square-edged, bell-mouth), Ghajar and Tam (1995) developed flow regime maps for the determination of the boundary between single-phase forced and mixed convection using experimental data of Ghajar and Tam (1994). Based on the ratio of the local peripheral heat transfer coefficient at the top and the bottom, the heat transfer data was classified as either forced or mixed convection among the different flow regimes. The forced-mixed convection boundary was then obtained by empirical correlations. From the flow maps, heat transfer correlations for different flow regimes were recommended. Recently Trafalis et al. (2005) used the Multiclass Support Vector Machines (SVM) method to classify vertical and horizontal two-phase flow regimes in 4 pipes with good accuracy. In this study, the SVM method was applied to the single-phase experimental data of Ghajar and Tam (1994) and new flow regime maps were developed. Five flow regimes (forced turbulent, forced transition, mixed transition, forced laminar, mixed laminar) were identified in the flow maps using Reynolds and Rayleigh numbers as the identifying parameters. The flow regimes on the boundaries of the new maps were represented by the SVM decision functions. The results show that the new flow regime maps for the three types of inlets can classify the forced and mixed convection experimental data in different flow regimes with good accuracy.

The combination of internet of things (IoT) with environmental sensing and image processing device has opened a new era to monitor the health of plants. Classification of plant diseases in early stages using image processing and analyzing environmental sensing data not only helps farmers to get healthy plants but also maximize the production. To monitor and classify plant diseases IoT is essential to send images and give feedback on it. In this paper, a raspberry pi based IoT device is proposed which sends images of plants to classify diseases and updates environmental parameters like air temperature, humidity, soil moisture and pH in MySQL database in real-time. To segment the affected part of plant, k-mean cluster algorithm is used after performing preprocessing stage and converting into L*a*b color space. Multi-class support vector machine (SVM) is applied to categorize disease using fourteen types of features of color, texture and shape obtained when implementing gray level co-occurrence matrix where the system was able to classify with an accuracy of 97.33%. Thus, classifying diseases and analyzing environment parameters help farms to monitor plant growth efficiently for better production.

Rice is the name of the type of plant that is needed by humans in the world. The plant is used as the main source of energy by Most people in the world, especially on the Asian continent. The importance of rice plants makes rice widely planted in various regions. Most humans use rice as a staple crop. Therefore, rice production needs to be considered to meet the need for enough food for most people in the world. The main thing that needs to be considered in maximizing rice production is that when guarding rice plants, many factors that inhibit rice plants can be the cause of food crises in various regions. Therefore, the care of rice production needs to be considered. In addition to the lack of nutrients in water and soil in decreasing rice production, plant diseases also need to be considered. Some types of diseases that often attack rice plants include bacterial leaf blight, brown spots, and left smut. Therefore, there is knowledge of prevention efforts or early treatment before the disease attacks rice plants more widely. The efficacy of technology can be used in solving this problem, we can take advantage of artificial intelligence in it. Artificial intelligence is implemented for the detection of types of diseases in rice plants using image images on the leaves of rice plants. If the disease in rice plants can be detected, it will make it easier for rice plant farmers to overcome the disease. The ANN (Artificial neural network) algorithm can be used in this problem from the results of research on identifying the type of rice disease using the algorithm obtained an accuracy of 83%. This shows the ability of artificial intelligence in disease identification can help farmers detect types of diseases.

The recognition and classification of rice plant diseases require an accurate system to generate classification data. Types of rice diseases can be identified in several ways, one of which is leaf characterization. One method that has high accuracy in identifying plant disease types is Convolutional Neural Networks (CNN). However, the rice disease data used has unbalanced data which affects the performance of the method. Therefore, the purpose of this research was to apply data augmentation to handle unbalanced rice disease data to improve the performance of the Convolutional Neural Network (CNN) method for rice disease type detection based on leaf images. The method used in this research is the CNN method for detecting rice disease types based on leaf images. The result of this research was the CNN method with 100 epochs able to produce an accuracy of 99.7% in detecting rice diseases based on leaf images with a division of 80% training data (2438 data) and 20% testing data (608 data). The conclusion is that the CNN method with the augmentation process can be used in rice disease detection because it has very high accuracy.

The agricultural sector is a vital part of the economy, providing food, raw materials, and employment opportunities. In Indonesia, this sector faces significant challenges, such as low interest from younger generations and plant disease issues. Plant disease identification typically requires experienced experts, but this process is time-consuming and costly. This research aims to develop a plant disease classification model using Convolutional Neural Network (CNN) to assist farmers in identifying diseases in rice, corn, tomato, and potato plants based on leaf images. Testing was conducted with data splitting ratios of 70:30, 80:20, and 90:10, using both single-stage and multi-stage classification methods. The best results were achieved with an 80:20 data ratio using single-stage classification, with an average accuracy of 80%, precision of 80%, recall of 81%, and F1-score of 79%. This study demonstrates that the CNN method is effective in plant disease classification, with optimal performance at an 80:20 data ratio and single-stage classification. It is hoped that this research can help farmers quickly and accurately identify and manage plant diseases, as well as encourage innovation in the agricultural sector. The implementation of CNN in plant disease classification shows great potential in enhancing the efficiency and accuracy of disease detection, ultimately supporting the sustainability and development of the agricultural sector.

In order to mitigate decreases in agricultural yield and production, the identification of diseases in plants assumes paramount importance. The agricultural sector has been employing various methodologies rooted in machine learning and image processing to address these challenges. This comprehensive analysis focuses specifically on the detection of diseases in rice plants by leveraging a diverse array of machine learning and image processing techniques with input images of infected rice plants. Furthermore, we delve into significant concepts pertaining to machine learning and image processing that aid in the identification and categorization of plant diseases. Various classification methods such as the k-Nearest Neighbor Classifier (KNN), Support Vector Machine (SVM), Probabilistic Neural Network (PNN), Genetic Algorithm (GA), and others find applications in agricultural research endeavors. The selection of an appropriate classification method assumes critical importance as the quality of the output is contingent on the input data. The classification of plant leaf diseases finds utility across multiple domains, including agriculture and biological research. This paper presents an extensive exploration of rice plant diseases, encompassing aspects such as image dataset size, preprocessing techniques, segmentation methods, and classifiers.

This paper aims to classify comments containing cyberbullying on instagram social media. Data were taken from the comments on instagram accounts of some well-known Indonesian Instagram celebrities/Selebgram, 1053 comments were taken as a training document and 34 comments were taken as a test document. Text classification method used is Support Vector Machine (SVM). The Support Vector Machine (SVM) method is used for classification types that only have two values, namely −1 and 1. In this classification process, the Support Vector Machine (SVM) method is used to see how far this method can classify comments on the Indonesian account Contains cyberbullying or not. The language used to create a text classification program with Support Vector Machine (SVM) is the R language using RStudio's Integrated Development Environment (IDE). The result shows that the classification of cyberbullying comments on instagram account of Indonesian program using Support Vector Machine (SVM) method shows the result of accuracy percentage of 79, 412%.