A Novel Deep Learning Framework Approach for Sugarcane Disease Detection

Abstract

Sugarcane, belonging to the grass family Poaceae, is rich in sugar sucrose, thereby used for making white sugar, jaggery and other by-products like molasses and bagasse. However, a diseased sugarcane plant is of no use, so it needs to be detected as soon as possible. A novel deep learning framework approach is proposed in this paper to detect whether a sugarcane plant is diseased or not by analyzing its leaves, stem, color, etc. The study comprises three scenarios based on different feature extractors namely Inception v3, VGG-16 and VGG-19. These are the pertained models on which different classifiers are trained. The state-of-the-art algorithms (SVM, SGD, ANN, naive Bayes, KNN and logistic regression) are compared with deep learning algorithms like neural network and hybrid AdaBoost. Several statistical measures such as accuracy, precision, specificity, AUC and sensitivity are calculated using Orange software, and the scenario having the highest accuracy is chosen. The receiver operating characteristic curve is computed in order to assess accuracy. An AUC of 90.2% is obtained using VGG-16 as the feature extractor and SVM as the classifier.

Riscos para a saúde do trabalhador e boas práticas de segurança do trabalho na aplicação de herbicidas em cana-de-açúcar

“…A cana-de-açúcar se caracteriza pelo seu sabor adocicado devido ao seu alto teor de sacarose. Destaca-se que uma planta adulta pode acumular até 25% de seu peso fresco na forma desse açúcar em condições normais de crescimento (Ansari et al, 2013), o que torna essa cultura a matéria-prima para a produção de cerca de 75% do açúcar consumido globalmente (Srivastava et al, 2020). Embora seja cultivada principalmente para a produção de açúcar, essa cultura produz inúmeros subprodutos de valor agregado, como melaço, bagaço e outros itens com destinação industrial para fabricação de produtos químicos, plásticos, tintas, sintéticos, fibras, inseticidas e detergentes (Mehnaz, 2013).…”

Prediction and Estimation of River Velocity Based on GAN and Multifeature Fusion

“…e classification model consists of one input layer and five convolutional processing units. Each unit contains two to three unequal convolution layers, a single pooling layer, two full connection layers, a Dropout layer, and a feature fusion layer [18][19][20][21][22]. e feature fusion mechanism is introduced to upsample the ow feature map output from convolution C5_2 layer.…”

Design and Experimental Evaluation of an Intelligent Sugarcane Stem Node Recognition System based on Enhanced YOLOv5s

“…Preeti Saini et al [17] proposed a novel hybrid method based on CNN-Bi-LSTM_CYP deep learning for predicting sugarcane crop yield, but this method is only capable of measuring yield using current data images. Srivastava S et al [18] proposed a novel network framework based on VGG and Inception for identifying sugarcane diseases and pests. Da Wang et al [19] employed the YOLOv5 network model to identify sugarcane internodes and designed a pre-cutting system for sugarcane varieties based on visual recognition technology, enabling cutting of different types of sugarcane.…”