International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 10 Issue: 05 | May 2023
p-ISSN: 2395-0072
www.irjet.net
Detection of Early Leaf spot of groundnut using Neural Network techniques Revati R. Nalawade1, S. D. Sawant2, P. M. Ingle3, V. G. More4 1Ph. D. Scholar, Department of Plant pathology, College of Agriculture, Dapoli 2Vice Chancellor, Dr. Balasaheb Sawant Konkan Krishi Vidyapeeth, Dapoli
3Associate Professor, Department of Irrigation & Drainage Engineering, CAET, Dapoli 4Agrometerologist, Department of Agronomy, College of Agriculture, Dapoli
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - India ranks second in groundnut and its oil
The major biotic factors affecting groundnut yield and quality are foliar diseases, viz. early (Cercospora arachidicola Hori.) and late leaf spots (Phaeoisariopsis personata Berk. And Curt.). These are the most widely distributed and economically important foliar disease of groundnut causing sever reduction in oil content. The combined losses due to both these leaf spots are more than 50% depending on the time of occurrence and congenial weather. The disease damage the plant by reducing the leaf area available for photosynthesis and stimulating the leaflet abscission leading to heavy defoliation [2].
production after China followed by USA and Nigeria (Tiwari et al., 2018). In Konkan region groundnut is grown on 20,000 ha area with a productivity of 1800 kg ha -1 [1]. The area under groundnut crop has increased enormously in Konkan region. The major biotic factors affecting groundnut yield and quality are foliar diseases, viz. early (Cercospora arachidicola Hori.) and late leaf spots (Phaeoisariopsis personata Berk. And Curt.). The combined losses due to both these leaf spots are more than 50% depending on the time of occurrence and congenial weather. The disease damage the plant by reducing the leaf area available for photosynthesis and stimulating the leaflet abscission leading to heavy defoliation [2]. In order to improve the recognition rate of disease diagnosis, researchers have studied many techniques using machine learning and pattern recognition such as Convolutional Neural Network, Artificial Neural Network, Back Propagation Neural Network, Support Vector Machine and other image processing methods. Due to higher performance capability in terms of computation and accuracy, Convolutional Neural Network (CNN) and Artificial Neural Network (ANN) models are most widely used for detection of plant diseases in agriculture [3]. With this view, the present investigation is planned to develop disease detection model with the help of Convolutional neural network and Artificial neural network for early leaf spot of groundnut caused by Cercospora arachidicola.
In order to improve the recognition rate of disease diagnosis, researchers have studied many techniques using machine learning and pattern recognition such as Convolutional Neural Network, Artificial Neural Network, Back Propagation Neural Network, Support Vector Machine etc. Due to higher performance capability in terms of computation and accuracy, Convolutional Neural Network (CNN) and Artificial Neural Network (ANN) models are most widely used for detection of plant diseases in agriculture [3]. With this view, the present investigation is planned to develop disease detection model with the help of Convolutional neural network and Artificial neural network for early leaf spot of groundnut caused by Cercospora arachidicola.
2. LITERATURE SURVEY
Key Words: Convolutional Neural network, Artificial
Neural network, Teachable machines, Mobiroller, Multilayer Perceptron, Conjugate Descent gradient, Levenberg Marquardt.
Kumar and Sowrirajan [5] proposed an image-processing based approach to automatically classify the normal or diseased leaves (Early leaf spot, Late leaf spot, Alternaria leaf spot). The RGB image samples of leaves of groundnut, mango, brinjal, tomato and maize were collected using high resolution camera. During pre-processing stage, the resizing of image to 256x256 pixels, color space conversion and region of interest selection was performed. Color, texture and geometric features of the image were extracted by the HSV conversion, GLCM and Lloyd’s clustering respectively. BPN-FF classifier was used for classification based on learning with the training samples and thereby provided the information on the disease (Early leaf spot, Late leaf spot and Alternaria leaf spot) as well as the respective control measures.
1.INTRODUCTION India ranks second in groundnut and its oil production after China followed by USA and Nigeria [4]. It contains 4850% oil and 26-28% protein, and a rich source of nutrients. In Konkan region groundnut is grown on 20,000 ha area with a productivity of 1800 kg ha-1 [1]. Optimum temperature and humidity with potash rich porous soil favors higher pod yield in groundnut in this region as compared to rest of the Maharashtra.
© 2023, IRJET
|
Impact Factor value: 8.226
|
ISO 9001:2008 Certified Journal
|
Page 1456