A novel hybrid inception-xception convolutional neural network for efficient plant disease classification and detection

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Jan. 31, 2025

Plants are essential at all stages of living things. Plant pests, diseases, and symptoms most regularly visible in plant leaves fruits sometimes within the roots. Yet, their diagnosis by experts laboratory is expensive, tedious, time-consuming if samples involve analysis. Failure to detect early diseases core biotic cause increased stresses, structure, health, reduced subsistence farming, threats global food security. To mitigate these problems a social, economic, environmental level, inappropriate herbicide application reduction disease detection classification (PDDC) significant solutions this case. Advancements transfer learning techniques have resulted effective results smart farming become extensively used identification research studies. This study presents novel hybrid inception-xception (IX) using convolution neural network (CNN). The presented model combines inception depth-separable layers capture multiple-scale features while reducing complexity overfitting. In contrast ordinary CNN architectures, it extends for better feature extraction, improving PDDC performance that demands diverse competencies. It further real-time artificial intelligence (AI) available MATLAB, Android, Servlet automatically identify classify based on leaf environment improved CNN, machine (ML), computer vision techniques. assess IX-CNN performance, different classifiers, namely, support vector (SVM), decision tree (DT) random forest (RF), were used. experiments six datasets, including PlantVillage, Turkey Disease, Doc, Rice RoCole, NLB datasets. Disease datasets demonstrated an accuracy 100%. attained 99.79%, 99.95%, 98.64%, respectively.

Language: Английский

---

Modern computational approaches for rice yield prediction: A systematic review of statistical and machine learning-based methods

Computers and Electronics in Agriculture, Journal Year: 2025, Volume and Issue: 231, P. 109852 - 109852

Published: Feb. 5, 2025

Language: Английский

---

Improving Remote Patient Monitoring and Care Using Machine Learning

Studies in computational intelligence, Journal Year: 2025, Volume and Issue: unknown, P. 179 - 193

Published: Jan. 1, 2025

Language: Английский

---

Machine Learning in Addiction Research: Advancements, Challenges, and Future Directions

Studies in computational intelligence, Journal Year: 2025, Volume and Issue: unknown, P. 207 - 219

Published: Jan. 1, 2025

Language: Английский

---

Introduction to Intelligent Techniques in Healthcare

Studies in computational intelligence, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 13

Published: Jan. 1, 2025

Language: Английский

---

Kuramoto Phase Model: Possible New Effects of Neuronal Dynamics

Studies in computational intelligence, Journal Year: 2025, Volume and Issue: unknown, P. 15 - 28

Published: Jan. 1, 2025

Language: Английский

---

Overcoming Challenges in the Integration of AI in Healthcare

Studies in computational intelligence, Journal Year: 2025, Volume and Issue: unknown, P. 283 - 300

Published: Jan. 1, 2025

Language: Английский

---

A Study on Performance of Ensemble Based Classifiers on Healthcare Data

Studies in computational intelligence, Journal Year: 2025, Volume and Issue: unknown, P. 43 - 60

Published: Jan. 1, 2025

Language: Английский

---

A novel early stage drip irrigation system cost estimation model based on management and environmental variables

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Feb. 3, 2025

One of the most significant, intricate, and little-discussed aspects pressurized irrigation is cost estimation. This study attempts to model early-stage drip system using a database 515 projects divided into four sections pumping station central control (TCP), on-farm equipment (TCF), installation operation (TCI), total (TCT). First, 39 environmental management features affecting listed sectors were extracted for each previously mentioned. A (a matrix × 43) was created, costs all updated baseline year 2022. Then, several feature selection algorithms, such as WCC, LCA, GA, PSO, ACO, ICA, LA, HTS, FOA, DSOS, CUK, employed choose significant that had biggest influence on cost. The carried out features) well easily available (those existed before system's design phase, 18 features). different machine learning models Multivariate Linear Regression, Support Vector Artificial Neural Networks, Gene Expression Programming, Genetic Algorithms, Deep Learning, Decision Trees, used estimate aforementioned sections. vector (SVM) optimization algorithms (Wrapper) found be best learner techniques, respectively, algorithms. two LCA FOA produced estimation, according evaluation criteria results. Their RMSE 0.0020 0.0018, their R2 0.94 0.94. For readily features, these 0.0006 0.95 both In part overall feature, modeling with selected revealed SVM (with RBF Kernel) among discussed. Its in training stage are = 0.923, 0.008, VE 0.082; testing stage, they 0.893, 0.009, 0.102. ANN (MLP) subset part, 0.912, 0.083 0.882, 0.103 stage. findings this can utilized highly accurately local systems based recognized parameters by employing particular models.

Language: Английский

---

Recognition of Multi-Symptomatic Rice Leaf Blast in Dual Scenarios by Using Convolutional Neural Networks

Smart Agricultural Technology, Journal Year: 2025, Volume and Issue: unknown, P. 100867 - 100867

Published: Feb. 1, 2025

Language: Английский

---