Application of Convoluted Neural Network and Its Architectures for Fungal Plant Disease Detection

2021 ◽  
pp. 314-324
Author(s):  
K. Bhargavi ◽  
B. Sathish Babu
Keyword(s):  
Neural Network ◽  
Plant Disease ◽  
Traditional Approach ◽  
Plant Diseases ◽  
Identification Accuracy ◽  
Disease Detection ◽  
Natural Ecosystem ◽  
Disease Identification ◽  
Fungal Plant Disease ◽  
Processing Techniques

Eighty-five percent of the plants are affected by diseases caused by organisms like fungus, bacteria, and virus, which devastate the natural ecosystem. The most common clues provided by the plants affected by fungal diseases are defaming of the plant color. In literature, several traditional rule-based algorithms and normal image processing techniques are used to identify the fungal plant diseases. However, the traditional approach suffers from poor disease identification accuracy. Convoluted neural network (CNN) is one of the potential deep learning neural networks used for image recognition and classification in plant pathology. In this chapter, some of the potential CNN architectures used for plant disease detection like LeNet, AlexNet, VGGNet, GoogLeNet, ResNet, and ZFnet are discussed with the architecture and advantages. The efficiencies achieved by ResNet and ZFNet are found to be good in terms of accuracy and error rate.

scholarly journals Tomato Leaf Disease Detection using Back Propagation Neural Network

2020 ◽  
Vol 9 (8) ◽  
pp. 529-538
Keyword(s):  
Neural Network ◽  
Traditional Approach ◽  
Back Propagation ◽  
Back Propagation Neural Network ◽  
Plant Diseases ◽  
Disease Detection ◽  
Large Drop ◽  
Tomato Crop ◽  
Leaf Disease ◽  
Area Of Interest

Most of the Indian economy rely on agriculture, so identifying any diseases crop in early stages is very crucial as these diseases in plants causes a large drop in the production and economy of the farmers and therefore, degradation of the crop which emphasize on the early detection of the plant disease. These days, detection of plant diseases has become a hot topic in the area of interest of the researchers. Farmers followed a traditional approach for identifying and detecting diseases in plants with naked eyes, which didn’t help much as the disease may have caused much damage to the plant. Tomato crop shares a huge portion of Indian cuisine and can be prone to various Air-Bourne and Soil-Bourne diseases. In this paper, we tried to automate the Tomato Plant Leaf disease detection by studying the various features of diseased and healthy leaves. The technique used is pattern recognition using Back-Propagation Neural network and comparing the results of this neural network on different features set. Several steps included are image acquisition, image pre-processing, features extraction, subset creation and BPNN classification.

Challenges and Issues in Plant Disease Detection Using Deep Learning

2021 ◽  
pp. 56-74
Author(s):  
Priyanka Sahu ◽  
Anuradha Chug ◽  
Amit Prakash Singh ◽  
Dinesh Singh ◽  
Ravinder Pal Singh
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Deep Learning ◽  
Plant Disease ◽  
Image Database ◽  
Plant Diseases ◽  
Disease Detection ◽  
Factors Influencing ◽  
Classification Tasks ◽  
Main Factors

Deep learning (DL) has rapidly become an essential tool for image classification tasks. This technique is now being deployed to the tasks of classifying and detecting plant diseases. The encouraging results achieved with this methodology hide many problems that are rarely addressed in related experiments. This study examines the main factors influencing the efficiency of deep neural networks for plant disease detection. The challenges discussed in the study are based on the literature as well as experiments conducted using an image database, which contains approximately 1,296 leaf images of the beans crop. A pre-trained convolutional neural network, EfficientNet B0, is used for training and testing purposes. This study gives and emphasizes on factors and challenges that may potentially affect the use of DL techniques to detect and classify plant diseases. Some solutions are also suggested that may overcome these problems.

scholarly journals Assessment of Plant Disease Identification using GLCM and KNN Algorithms

2020 ◽  
Vol 8 (5) ◽  
pp. 4900-4904
Keyword(s):  
Machine Learning ◽  
Classification Accuracy ◽  
Plant Disease ◽  
Nearest Neighbors ◽  
Input Image ◽  
Plant Diseases ◽  
Disease Detection ◽  
Textural Features ◽  
K Nearest Neighbors ◽  
Disease Identification

One of the significant segments of Indian Economy is Cultivation. Occupation to almost 50% of the nation’s labor force is delivered by Indian cultivation segment. India is recognized to be the world's biggest manufacturer of pulses, rice, wheat, spices and spice harvests. Agronomist's financial progress is contingent on the excellence of the goods that they yield, which depend on on the plant's progress and the harvest they get. Consequently, in ground of cultivation, recognition of disease in plants shows an involved part. Plants are exceedingly disposed to to infections that disturb the progress of the plant which in chance distresses the natural balance of the agronomist. In order to distinguish a plant disease at right preliminary period, usage of automatic disease detection procedure is beneficial. The indications of plant diseases are noticeable in various portions of a plant such as leaves, etc. Physical recognition of plant disease by means of leaf descriptions is a wearisome job. The k-mean clustering procedure is utilized for the segmentation of input images. The GLCM (gray-level co-occurrence matrices) procedure is utilized which excerpts textural features from the input image and implementation of KNN (k-nearest neighbors) algorithm for image classification and produced classification accuracy from 70 to 75% for different inputs. Hence, it is required to develop machine learning based computational methods which will make the process of disease detection and classification using leaf images automatic. .. To advance concert of standing methods machine learning and deep learning algorithms will be utilized for more accurate classification.

scholarly journals MACHINE LEARNING TECHNIQUES IN PLANT DISEASE DETECTION AND CLASSIFICATION – A STATE OF THE ART

2021 ◽  
pp. 362-372
Author(s):  
John Sreya ◽  
Leena Rose Arul
Keyword(s):  
Machine Learning ◽  
Plant Disease ◽  
Machine Learning Algorithms ◽  
Plant Diseases ◽  
Machine Learning Techniques ◽  
Small Scale ◽  
Disease Detection ◽  
Detection Systems ◽  
Learning Techniques ◽  
Processing Techniques

As we belong to a developing country, the agricultural importance is a known criterion. Majority of the Indians depend on agriculture for their basic living. It also serves as the backbone of the Indian economy. Therefore this sector should be considered important and taken care of. Diseases affecting the plants and pest are the two major threats of agriculture production. Naked eye observation followed by the addition of chemical fertilizers is the traditional method adopted by most of the farmers to avoid plant diseases. But the main limitation to this method is that it works only in the case of small scale farming. In order to tackle this issue many automatic plant disease detection systems have been developed from the early 70s. This paper is intended to survey some of the existing works in plant disease recognition that include various procedures, materials and approaches. They use different machine learning algorithms, image processing techniques and deep learning methods for disease detection. This paper also compares and suggests novel methods to recognize and classify the various kinds of infections affecting agricultural plants.

Deep Learning-Based Knowledge Extraction From Diseased and Healthy Edible Plant Leaves

2021 ◽  
Vol 12 (2) ◽  
pp. 67-81
Author(s):  
Udit Jindal ◽  
Sheifali Gupta
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Plant Disease ◽  
Disease Detection ◽  
Edible Plant ◽  
Research Article ◽  
Segmented Images ◽  
Significant Area ◽  
Testing Accuracy ◽  
Disease Present

Agriculture contributes majorly to all nations' economies, but crop diseases are now becoming a very big issue that has to be resolving immediately. Because of this, crop/plant disease detection becomes a very significant area to work. However, a huge number of studies have been done for automatic disease detection using machine learning, but less work has been done using deep learning with efficient results. The research article presents a convolution neural network for plant disease detection by using open access ‘PlantVillage' dataset for three versions that are colored, grayscale, and segmented images. The dataset consists of 54,305 images and is being used to train a model that will be able to detect disease present in edible plants. The proposed neural network achieved the testing accuracy of 99.27%, 98.04%, and 99.14% for colored, grayscale, and segmented images, respectively. The work also presents better precision and recall rates on colored image datasets.

scholarly journals A Review on Advances in Automated Plant Disease Detection

2021 ◽  
Vol 11 (4) ◽  
pp. 251-264
Author(s):  
Radhika Bhagwat ◽  
Yogesh Dandawate
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Plant Disease ◽  
Learning Algorithms ◽  
Machine Learning Algorithms ◽  
Plant Diseases ◽  
Detection Methods ◽  
Economic Losses ◽  
Disease Detection ◽  
Range Images

Plant diseases cause major yield and economic losses. To detect plant disease at early stages, selecting appropriate techniques is imperative as it affects the cost, diagnosis time, and accuracy. This research gives a comprehensive review of various plant disease detection methods based on the images used and processing algorithms applied. It systematically analyzes various traditional machine learning and deep learning algorithms used for processing visible and spectral range images, and comparatively evaluates the work done in literature in terms of datasets used, various image processing techniques employed, models utilized, and efficiency achieved. The study discusses the benefits and restrictions of each method along with the challenges to be addressed for rapid and accurate plant disease detection. Results show that for plant disease detection, deep learning outperforms traditional machine learning algorithms while visible range images are more widely used compared to spectral images.

Optimized fuzzy neural network for Jatropha Curcas plant disease identification

2017 ◽  
Author(s):  
Diny Melsye Nurul Fajri ◽  
Triando Hamonangan Saragih ◽  
Andi Hamdianah ◽  
Wayan Firdaus Mahmudy ◽  
Yusuf Priyo Anggodo
Keyword(s):  
Neural Network ◽  
Jatropha Curcas ◽  
Fuzzy Neural Network ◽  
Plant Disease ◽  
Disease Identification ◽  
Fuzzy Neural

Plant Disease Detection Using Machine Learning Approaches

2021 ◽  
pp. 122-130
Author(s):  
Sukanta Ghosh ◽  
Shubhanshu Arya ◽  
Amar Singh
Keyword(s):  
Plant Disease ◽  
Crop Yields ◽  
Detection System ◽  
Visual Observation ◽  
Plant Diseases ◽  
Disease Detection ◽  
Learning Approaches ◽  
Factors Affecting ◽  
Plant Leaf ◽  
Market Situation

Agricultural production is one of the main factors affecting a country's domestic market situation. Many problems are the reasons for estimating crop yields, which vary in different parts of the world. Overuse of chemical fertilizers, uneven distribution of rainfall, and uneven soil fertility lead to plant diseases. This forces us to focus on effective methods for detecting plant diseases. It is important to find an effective plant disease detection technique. Plants need to be monitored from the beginning of their life cycle to avoid such diseases. Observation is a kind of visual observation, which is time-consuming, costly, and requires a lot of experience. For speeding up this process, it is necessary to automate the disease detection system. A lot of researchers have developed plant leaf detection systems based on various technologies. In this chapter, the authors discuss the potential of methods for detecting plant leaf diseases. It includes various steps such as image acquisition, image segmentation, feature extraction, and classification.

Sign in / Sign up

Export Citation Format

Share Document