Crop disease detection via ensembled-deep-learning paradigm and ABC Coyote pack optimization algorithm (ABC-CPOA)

Crop disease is a significant issue that affects the growth and yield of crops, leading to financial loss for farmers. Identification and treatment of crop diseases have become challenging due to the increase in the variety of diseases and the lack of knowledge among farmers. To address this issue,...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Multimedia tools and applications Ročník 84; číslo 1; s. 37 - 62
Hlavní autori: Chithambarathanu, M., Jeyakumar, M. K.
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: New York Springer US 01.01.2025
Springer Nature B.V
Predmet:
Algorithms
Artificial neural networks
Color
Computer Communication Networks
Computer Science
Convexity
Crop diseases
Data Structures and Information Theory
Deep learning
Feature extraction
Heuristic methods
Image filters
Machine learning
Medical imaging
Multimedia Information Systems
Neural networks
Optimization algorithms
Optimization models
Plant diseases
Recurrent neural networks
Shape
Special Purpose and Application-Based Systems
Weighting functions
O-DNN
Crop disease detection
RNN
Attribute based -Bi LSTM
Self-Improved Coyote Optimization Algorithm
ISSN:1573-7721, 1380-7501, 1573-7721
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:Crop disease is a significant issue that affects the growth and yield of crops, leading to financial loss for farmers. Identification and treatment of crop diseases have become challenging due to the increase in the variety of diseases and the lack of knowledge among farmers. To address this issue, this investigate uses an ensembled-deep-learning paradigm to propose a deep learning-based model for crop disease identification trained with an ABC-CPOA. Initially, collected raw images are pre-processed via Bilateral filter and gamma correction Feature Extraction: Then, from the pre-processed images, the features like texture feature (Local Quinary Pattern (LQP), Local Gradient Pattern (LGP), Enriched Local Binary Pattern (E-LBP), color features (Color Histogram, Color Moments), shape features (Contour-based features, Convex Hull). Optimal feature selection- Among the extracted features, the optimal features is designated by means of a self-improved meta-heuristic optimization model referred as ABC-CPOA. This ABC-CPOA model is an extended version of standard Coyote Optimization Algorithm (COA). Crop disease detection phase is modelled with a new ensembled-deep-learning paradigm. Ensembled-deep-learning paradigm comprises Attention-based Bi-LSTM, Recurrent Neural Networks (RNNs) and Optimized Deep Neural Network (O-DNN). The weight function of O-DNN is fine-tuned using the new ABC-CPOA. Precision, recall, sensitivity, and specificity, in addition to TPR, FPR, FNR, and TNR, F1-score, and accuracy are used to assess the suggested approach. The implementation was performed by the MATLAB tool (version: 2022B).
Bibliografia:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1573-7721
1380-7501
1573-7721
DOI:10.1007/s11042-024-19329-y