Identification Of Plant Leaf Diseases Research Articles

Plant leaf Disease Identification using Deep learning techniques

Agriculture is an important source of our country’s growth. The major loss in an agricultural economy is because of the plant disease. Though technology plays a vital role in all the fields still today the agriculture field is using the old methodologies. Successful cultivation depends on identifying plant disease. Previously the identification was done manually by the experienced people but now it became difficult due to environmental changes. By using the deep learning techniques the plant disease can be identified effectively. Vgg16 and ResNet are the proposed techniques to increase accuracy than the existing system. The disease can be identified with images of the leaves by applying those deep learning techniques. Detection can be involved in steps like image acquisition, image pre-processing, image segmentation, feature extraction, and classification. By controlling the disease, productivity can be increased. The features of the leaf image are taken and trained using the neural network algorithm and then the prediction is done by testing the images. The features of the leaf image are taken and trained using the neural network algorithm and then the prediction is done by testing the images.

Research on Plant Leaf Disease Identification Based on Transfer Learning Algorithm

Plant disease is one of the important threat factors that hinder the normal growth and development of plants. The intelligent identification of plant disease species has become increasingly important in the agricultural field. In This paper, the open-source data set including Black rot, bacterial spot, rust, and healthy leaves are used to train the ResNet model. And the transfer learning algorithm is applied on ResNet to establish a plant disease recognition model with good versatility and high training efficiency. The experiment results show that the disease identification accuracy of the transfer learning model is 83.75%, which is much higher than that of the traditional ResNet-101 model. Therefore, the plant disease recognition model based on transfer learning algorithm is highly feasible.

Using deep transfer learning for image-based plant disease identification

Plant diseases have a disastrous impact on the safety of food production, and they can cause a significant reduction in both the quality and quantity of agricultural products. In severe cases, plant diseases may even cause no grain harvest entirely. Thus, the automatic identification and diagnosis of plant diseases are highly desired in the field of agricultural information. Many methods have been proposed for solving this task, where deep learning is becoming the preferred method due to the impressive performance. In this work, we study transfer learning of the deep convolutional neural networks for the identification of plant leaf diseases and consider using the pre-trained model learned from the typical massive datasets, and then transfer to the specific task trained by our own data. The VGGNet pre-trained on ImageNet and Inception module are selected in our approach. Instead of starting the training from scratch by randomly initializing the weights, we initialize the weights using the pre-trained networks on the large labeled dataset, ImageNet. The proposed approach presents a substantial performance improvement with respect to other state-of-the-art methods; it achieves a validation accuracy of no less than 91.83% on the public dataset. Even under complex background conditions, the average accuracy of the proposed approach reaches 92.00% for the class prediction of rice plant images. Experimental results demonstrate the validity of the proposed approach, and it is accomplished efficiently for plant disease detection.

Plant leaf disease detection and control: A survey

The most important aspect of the Indian economy is agriculture. It is very common and natural to have a disease in plants with varying climatic conditions. This further leads to the crop quality getting deteriorated. With the recent changes in the weather cycles, achieving the best quality and quantity of crop is the most challenging task for farmers. With such challenges at the forefront, image processing has been proved as the best technique to detect the initial stages of disease based on the color, texture, and shape of the crop leaf. There are many different feature extraction techniques like color histogram, canny and Sobel edge detector, gray level co-occurrence matrix, Gabor filter which are used for extracting the feature of the disease in a crop leaf. Once disease features are extracted then classification algorithms like Support Vector Machine, Artificial Neural Network, Backpropagation, Convolution Neural Network, Feed Forward Neural Network, Probabilistic Neural Network, and Radial Neural Network are used to classify the disease. Once disease reason is identified, then proper treatment can be applied after identifying the reason behind the disease. These image processing and classification techniques have been proved accurate but feature extraction is the most time-consuming method as it is done manually using the different methods. Also, image processing with classification works for the small dataset. Convolutional Neural Networks as a part of deep learning, and it’s a most effective sub branch of image processing. Many applications for the automatic identification of plant leaf diseases have been developed. These applications could serve as a basis for the development of expert assistance. Such types of tools could contribute great and sustainable agricultural practices and greater food production security. To examine the potential of these networks for such applications, we survey research studies that relied on CNNs to automatically identify plant leaf diseases. We describe their performance and main implementation aspects. Our survey allows us to identify the issues in this research area. This work covers the survey of different types of the disease occur in plants, its causes, symptoms, and treatment policy with the image processing and classification methods used. This survey also discussed the limitations of existing research work which provides the direction for further improvements in plant disease identification and preventive measures.

Identification of plant leaf diseases using a nine-layer deep convolutional neural network

In this paper, we proposed a novel plant leaf disease identification model based on a deep convolutional neural network (Deep CNN). The Deep CNN model is trained using an open dataset with 39 different classes of plant leaves and background images. Six types of data augmentation methods were used: image flipping, gamma correction, noise injection, principal component analysis (PCA) colour augmentation, rotation, and scaling. We observed that using data augmentation can increase the performance of the model. The proposed model was trained using different training epochs, batch sizes and dropouts. Compared with popular transfer learning approaches, the proposed model achieves better performance when using the validation data. After an extensive simulation, the proposed model achieves 96.46% classification accuracy. This accuracy of the proposed work is greater than the accuracy of traditional machine learning approaches. The proposed model is also tested with respect to its consistency and reliability.

SLIC_SVM based leaf diseases saliency map extraction of tea plant

For the purpose of improving the extraction of tea plant leaf disease saliency map under complex backgrounds, a new algorithm combining SLIC (Simple Linear Iterative Cluster) with SVM (Support Vector Machine) is proposed in this paper. Firstly, super-pixel block is obtained by SLIC algorithm, significant point is detected by Harris algorithm, and fuzzy salient region contour is extracted by employing convex hull method. Secondly, the four-dimensional texture features of super-pixel blocks in salient regions and background areas are extracted, and then the classification map is obtained by classifying the super-pixel blocks with the help of SVM classifier. Lastly, the morphological and algebraic operations are implemented for repairing classified super-pixel blocks. As a result, one accurate saliency map of tea plant leaf disease image is obtained. Through testing based on 261 diseased images, the quality evaluation index, the accuracy, precision, recall and F-value are 98.5%, 96.8%, 98.6% and 97.7%, respectively. It demonstrates that the proposed method performs better than the other three SLIC-based algorithms in visual effects and quality assessment index. Such conclusion can be drawn that the proposed method can effectively extract tea plant leaf disease saliency map from complex background. Consequently, this research is expected to lay a good basis for the study of tea plant leaf disease identification. Last but not the least, the proposed method has good potential that extracts saliency map of crops or plants disease.

Plant leaf disease identification using exponential spider monkey optimization

Agriculture is one of the prime sources of economy and a large community is involved in cropping various plants based on the environmental conditions. However, a number of challenges are faced by the farmers including different diseases of plants. The detection and prevention of plant diseases are the serious concern and should be treated well on time for increasing the productivity. Therefore, an automated plant disease detection system can be more beneficial for monitoring the plants. Generally, the most diseases may be detected and classified from the symptoms appeared on the leaves. For the same, extraction of relevant features plays an important role. A number of methods exists to generate high dimensional features to be used in plant disease classification problem such as SPAM, CHEN, LIU, and many more. However, generated features also include unrelated and inessential features that lead to degradation in performance and computational efficiency of a classification problem. Therefore, the choice of notable features from the high dimensional feature set is required to increase the computational efficiency and accuracy of a classifier. This paper introduces a novel exponential spider monkey optimization which is employed to fix the significant features from high dimensional set of features generated by SPAM. Furthermore, the selected features are fed to support vector machine for classification of plants into diseased plants and healthy plants using some important characteristics of the leaves. The experimental outcomes illustrate that the selected features by Exponential SMO effectively increase the classification reliability of the classifier in comparison to the considered feature selection approaches.