An Improved Deep Residual Convolutional Neural Network for Plant Leaf Disease Detection

Abstract

In this research, we proposed a novel deep residual convolutional neural network with 197 layers (ResNet197) for the detection of various plant leaf diseases. Six blocks of layers were used to develop ResNet197. ResNet197 was trained and tested using a combined plant leaf disease image dataset. Scaling, cropping, flipping, padding, rotation, affine transformation, saturation, and hue transformation techniques were used to create the augmentation data of the plant leaf disease image dataset. The dataset consisted of 103 diseased and healthy image classes of 22 plants and 154,500 images of healthy and diseased plant leaves. The evolutionary search technique was used to optimise the layers and hyperparameter values of ResNet197. ResNet197 was trained on the combined plant leaf disease image dataset using a graphics processing unit (GPU) environment for 1000 epochs. It produced a 99.58 percentage average classification accuracy on the test dataset. The experimental results were superior to existing ResNet architectures and recent transfer learning techniques.

An Intelligent Approach for Cotton Plant Disease Detection using Convolutional Neural Networks: A Deep Learning Perspective

“…The study [21] introduced a new and improved computer program called ResNet197 to identify various diseases in plant leaves. This ResNet197 has 197 layers and is quite powerful.…”

Optimized deep learning-based dual segmentation framework for diagnosing health of apple farming with the internet of things

“…Fitness function cannot adapt to potential change when test dataset is increased K-Means clustering, color segmentation [19] Simplified approach, effective segmentation Narrowed analysis to offer higher precision SVM, KNN [20] Supports instance-based learning Biased prediction not addressed; outcomes have outliers too Swarm Optimization, FCM [21] Applicable for varied forms of images even with noisy Highly sensitive during initialization & parameters Wavelet, ANN [22] -Simplified feature learning for leaf disease -Demands large number of parameters, leading to overfitting SVM [23] -Capable of processing highdimensional problem space -Less overfitting issues -Computationally intensive operation when exposed to multiple classes of leaf disease MobileNet [25] -Faster inference -Effective transfer learning and finetuning -Reduced model capacity -Sensitive to input quality -Lower accuracy CNN with ResNet [27] Optimal resource dependencies -Higher computational burden -Higher accuracy Proposed (CNN, Genetic Algorithm, U-Net) -Lower computational burden -Higher accuracy -Highly optimized performance -Supports generalization -Highly adaptable and flexible -Not assessed over high resolution images…”

Grey Blight Disease Detection on Tea Leaves Using Improved Deep Convolutional Neural Network

“…Tis approach relies on the extraction of features from the leaf images and their identifcation and classifcation using an artifcial neural network (ANN) [8]. Deep learning, an advanced machine learning technique, which uses deep convolutional neural networks (DCNN) for crop disease identifcation, is gaining increased application due to its automatic feature extraction ability, accuracy, and robustness in detection [9][10][11]. For tea disease detection and classifcation, a few machine learning-based approaches have been employed with considerable performance [5,[12][13][14][15].…”