Browsing by Author "Sharif, Md. Shahin"

Now showing 1 - 2 of 2

Jute Leaf Disease Prediction Using Deep Neural Network
(Daffodil International University, 2018-12-09) Islam, Md. Ariful; Sharif, Md. Shahin; Kafi, Md. Abdullahel; Arefin, Md. Shamsul
Deep constitutional neural network is a diverting area where research researches and achievement are taking geometrical progress in the agriculture field. Various researches are going on vigorously in plant diseases detection. Plant contribution is highly important for human life and environment. Plants also suffer from diseases as human and animals. There are many plant diseases that occurs and affected natural growth of plant. These diseases infected complete plants including leaf, stem, root, fruit and flowers. This research propose is a diseases detection and classification technique with the help of Deep learning convolution neural network. The latest generation of constitutional neural networks (CNN's) has gained magnificent results in the field of image classification. This research is related with a new approach to the development of plant disease detection model, based on leaf image classification, by using deep constitutional neural networks (DCNN). All essential steps required for implementing this disease detection model is fully discussed throughout the report, starting from collecting images in order to generate a dataset, evaluated by agricultural experts. This research is mainly focused how to detect jute plant leaves diseases using training and testing data using Deep learning convolution neural network (DCNN), stable and data set.
Sugarcane Stem and Leaf Disease Prediction Using Deep Neural Network
(Daffodil International University, 2020-07-09) Sharif, Md. Shahin
Deep convolutional neural network is a diverting area where researches and achievement are taking excellent progress in agriculture field. The most recent enhancements in computer vision formulated thorough deep learning have covered the method for how to identify and analyze diseases in plants by utilizing a camera to capture images an basis for recognizing several types of plant diseases. This research elaborates disease detection and classification with help of deep learning convolutional neural network. Sugarcane is a vital crop in the world. For detecting sugarcane diseases the researchers used the convolutional neural networks (CNNs) as the basic deep learning method. This study trained and test deep learning model consisting of 2200 sugarcane images dataset. After applying CNNs it achieves an accuracy of 92% and also get error rate 8%. The trained model acquired it motive by detecting and classifying sugarcane images into healthy and infected of sugarcane plants. Therefore, this research provides a step of helping farmers with the process of deep learning algorithm in detecting and classifying sugarcane diseases.