Image Recognition of Peanut Leaf Diseases Based on Capsule Networks

Dong, Mengping; Mu, Shaomin; Su, Tingting; Sun, Weiyuan

doi:10.1007/978-981-32-9298-7_4

Cited by 9 publications

(5 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(Hang et al, 2019) combined the structures of VGG16 (as the base), Squeeze and Ex-citation (SE) network and Inception module, replacing FC layer with global average pooling, which resulted into a reduced number of parameters and better classification accuracy. (Dong et al, 2019) utilized Capsule nAtworks (CapsNet) (Sabour et al, 2017) for disease identification in peanut crop, whereby three convolutional layers were placed with the original CapsNet and performances compared with a standard CNN. CapsNets have also been implemented for hyper-spectral image classification (Paoletti et al, 2018) and identification of UAV imagery (Li et al, 2019).…”

Section: Related Workmentioning

confidence: 99%

SE-CapsNet: Automated evaluation of plant disease severity based on feature extraction through Squeeze and Excitation (SE) networks and Capsule networks

Verma

Jahangir²,

Chug³

et al. 2021

KJS publishes peer-review articles in Mathematics, Computer Sci

View full text Add to dashboard Cite

Diseases in plants harm the quantity of the overall food production as well as the quality of the yield. Early detection, diagnosis and treatment can greatly reduce losses, both economic and ecological. Intuitively, reduction in the use of agrochemicals due to timely detection of the disease, would greatly help in mitigating the environmental impact. In this paper, the authors have proposed an improved feature computation approach based on Squeeze and Excitation (SE) Networks, before processing by the original Capsule networks (CapsNet) for classification, for estimating the disease severity in plants. Two SE networks, one based on AlexNet and another on ResNet have been combined with Capsule networks. Leaf images for the devastating Late Blight disease occurring in the Tomato crop have been utilized from the PlantVillage dataset. The images, divided into four severity stages i.e. healthy, early, middle and end, are downscaled, enhanced and given as input to the SE networks. The feature maps generated from the two networks are separately given as input to the Capsule Network for classification and their performances are compared with the original CapsNet, on two image sizes 32X32 and 64X64. SE-Alex-CapsNet achieves the highest accuracy of 92.1% and SE-Res CapsNet achieves the highest accuracy of 93.75% with 64X64 image size, as compared to CapsNet that results in 85.53% accuracy. The classification accuracies of six state-of-the-art CNN models namely AlexNet, SqueezeNet, ResNet50, VGG16, VGG19 and Inception V3 are also presented for comparison purposes. Accuracy as well as precision, recall, F1-score, validation loss etc. measures have been recorded and compared. The findings have been validated by implementing the proposed approaches with another dataset, achieving similar resultant accuracy measures. The implementation was also accomplished with datasets after noise addition in six different variations, to verify the robustness of the proposed model. Based on the performances, the proposed techniques can be exploited for disease severity assessment in other crops as well and can be extended to other areas of applications such as plant species classification, weed identification etc. In addition to improved performance, with reduced image size, the proposed methodology can be utilized to create a mobile application requiring low processing capabilities, to be installed on reasonably priced smartphones for practical usage by farmers.

show abstract

Section: Related Workmentioning

confidence: 99%

SE-CapsNet: Automated evaluation of plant disease severity based on feature extraction through Squeeze and Excitation (SE) networks and Capsule networks

Verma

Jahangir²,

Chug³

et al. 2021

KJS publishes peer-review articles in Mathematics, Computer Sci

View full text Add to dashboard Cite

show abstract

“…Due to its structure, training a deep convolutional neural network requires high computing power [12] and a large amount of data in order to attain better results [13]. This technology, however, is not yet used, explored, and examined in the case of peanut farming [14], particularly in the identification of peanut leaf spot disease due to aspects such as the availability of sufficient data needed to train and test the model [11].…”

Section: Introductionmentioning

confidence: 99%

Peanut leaf spot disease identification using pre-trained deep convolutional neural network

Patayon

Crisostomo

2022

IJECE

View full text Add to dashboard Cite

<span>Reduction of quality and quantity of agricultural products, particularly peanut or groundnut, is usually associated with disease. This could be solved through automatic identification and diagnoses using deep learning. However, this technology is not yet explored and examined in the case of peanut leaf spot disease due to some aspects, such as the availability of sufficient data to be used for training and testing the model. This study is intended to explore the use of pre-trained visual geometry group–16 (VGG16), visual geometry group–19 (VGG19), InceptionV3, MobileNet, DenseNet, Xception, InceptionResNetV2, and ResNet50 architectures and deep learning optimizers such as stochastic gradient descent (SGD) with Momentum, adaptive moment estimation (Adam), root mean square propagation (RMSProp), and adaptive gradient algorithm (Adagrad) in creating a model that can identify leaf spot disease by using a total of 1,000 images of leaves captured using a mobile camera. Confusion matrix was used to assess the accuracy and precision of the results. The result of the study shows that DenseNet-169 trained using SGD with momentum, Adam, and RMSProp attained the highest accuracy of 98%, while DenseNet-169 trained using RMSProp achieved the highest precision of 98% among pre-trained deep convolutional neural network architectures. Furthermore, this result could be beneficial in agricultural automation and disease identification systems for peanut or groundnut plants.</span>

show abstract

“…Initially, RGB was converted into HSV and then plane separation and color features extraction steps were carried out. Dong et al (2019) [9] applied a capsule network for peanut leaf disease recognition with the use of dynamic routing to overcome the problem of rotational invariance and spatial relationships. Their empirical observations showed that the recognition accuracy of the capsule network is 82.17% which is better than the *Corresponding Author.…”

Section: Introductionmentioning

confidence: 99%

Groundnuts Leaf Disease Recognition using Neural Network with Progressive Resizing

Rakholia¹,

Tailor²,

Saini³

et al. 2022

IJACSA

View full text Add to dashboard Cite

Groundnut is an important oilseed crop in the world, and India is the second-largest producer of groundnuts. This crop is prone to attack by numerous diseases which is one of the most important factors contributing to the loss of productivity and degradation in the quality; both of these finally result in a low agricultural economy. Therefore, it is necessary to find better and more reliable automation solutions to recognize groundnut leaf diseases. In this paper, a deep learning based model with progressive resizing is proposed for groundnut leaf disease recognition and classification tasks. Five major categories of groundnut leaf diseases namely leaf spot, armyworms effect, wilts, yellow leaf, and healthy leaf are considered. The proposed model was trained with and without progressive resizing while it was validated using cross-entropy loss. The first of its kind dataset used for training and validation purposes was manually created from the Saurashtra region of Gujarat state of India. The created dataset was imbalanced in terms of a different number of samples for each category. To handle the imbalanced dataset problem, the extended focal loss function was used. To evaluate the performance of the proposed model, different performance measures including precision, sensitivity, F1-score, and accuracy were applied. The proposed model achieved state-of-the-art accuracy of 96.12%. The model with progressive resizing performed better than the traditional core neural network-based model built on cross-entropy loss.

show abstract

Image Recognition of Peanut Leaf Diseases Based on Capsule Networks

Cited by 9 publications

References 3 publications

SE-CapsNet: Automated evaluation of plant disease severity based on feature extraction through Squeeze and Excitation (SE) networks and Capsule networks

SE-CapsNet: Automated evaluation of plant disease severity based on feature extraction through Squeeze and Excitation (SE) networks and Capsule networks

Peanut leaf spot disease identification using pre-trained deep convolutional neural network

Groundnuts Leaf Disease Recognition using Neural Network with Progressive Resizing

Contact Info

Product

Resources

About