Advancing Legume Quality Assessment Through Machine Learning: Current Trends and Future Directions

Journal of Food Composition and Analysis, Journal Year: 2025, Volume and Issue: unknown, P. 107532 - 107532

Published: March 1, 2025

Language: Английский

---

Deep learning and evolutionary intelligence with fusion-based feature extraction for classification of wheat varieties

European Food Research and Technology, Journal Year: 2025, Volume and Issue: unknown

Published: April 30, 2025

Language: Английский

---

Maize seeds forecasting with hybrid directional and bi‐directional long short‐term memory models

Food Science & Nutrition, Journal Year: 2023, Volume and Issue: 12(2), P. 786 - 803

Published: Nov. 9, 2023

The purity of the seeds is one important factors that increase yield. For this reason, classification maize cultivars constitutes a significant problem. Within scope study, six different models were designed to solve A special dataset was created be used in for study. contains total 14,469 images four classes. Images belong types, BT6470, CALIPOS, ES_ARMANDI, and HIVA, taken from BIOTEK company. AlexNet ResNet50 architectures, with transfer learning method, image classification. In order improve success, LSTM (Directional Long Short-Term Memory) BiLSTM (Bi-directional algorithms architectures hybridized. As result classifications, highest success obtained ResNet50+BiLSTM model 98.10%.

Language: Английский

---

Classification of hazelnut varieties based on bigtransfer deep learning model

European Food Research and Technology, Journal Year: 2024, Volume and Issue: 250(5), P. 1433 - 1442

Published: Feb. 27, 2024

Abstract Hazelnut is an agricultural product that contributes greatly to the economy of countries where it grown. The human factor plays a major role in hazelnut classification. typical approach involves manual inspection each sample by experts, process both labor-intensive and time-consuming, often suffers from limited sensitivity. deep learning techniques are extremely important classification detection products. Deep has great potential sector. This technology can improve quality, increase productivity, offer farmers ability classify detect their produce more effectively. for sustainability efficiency industry. In this paper aims application algorithms streamline classification, reducing need labor, time, cost sorting process. study utilized images three different varieties: Giresun, Ordu, Van, comprising dataset 1165 1324 1138 Van hazelnuts. open-access dataset. study, experiments were carried out on determination varieties with BigTransfer (BiT)-M R50 × 1, BiT-M R101 3 R152 4 models. models, including big transfer was employed task involved 3627 nut resulted remarkable accuracy 99.49% model. These innovative methods also lead patentable products devices various industries, thereby boosting economic value country.

Language: Английский

---

Deep Learning-Based Classification of Black Gram Plant Leaf Diseases: A Comparative Study

Proceedings of the International Conference on Advanced Technologies, Journal Year: 2023, Volume and Issue: unknown

Published: Aug. 19, 2023

The escalating incidence of plant diseases presents considerable obstacles to the agricultural domain, resulting in substantial reductions crop yield and posing a threat food security. To address pressing concern Black Gram Plant Leaf Diseases (BPLD), this research endeavors tackle disease classification through application deep learning methodology. approach leverages comprehensive dataset that encompasses Anthracnose, Crinkle, Powdery Mildew, Yellow Mosaic diseases, all which affect black gram crop. By employing advanced technique, we aim contribute valuable insights combat BPLD effectively. Our applies models, including Darknet-53, ResNet-101, GoogLeNet, EfficientNet-B0, classify diseases. Darknet-53 achieved 98.51% accuracy, followed by ResNet-101 (97.51%), GoogLeNet (96.52%), EfficientNet-B0 (77.61%). These findings demonstrate potential for accurate identification, benefiting agriculture. study provides comparative analysis models Disease (BPLD) classification, revealing as superior performers. Implementing these real-world scenarios holds promise early detection intervention, reducing losses. high accuracy signifies significant progress automating recognition, sector.

Language: Английский

---

Automated classification of hand-woven and machine-woven carpets based on morphological features using machine learning algorithms

Journal of the Textile Institute, Journal Year: 2024, Volume and Issue: 115(12), P. 2650 - 2659

Published: Feb. 11, 2024

As a cultural heritage, hand-woven carpets engender trust and admiration in individuals who recognize their authenticity. It is the expertise of experts determine whether carpet or machine-woven based on authenticity criteria. A total 48 morphological features were extracted by three from 359 handwoven machine woven carpets. Machine-learning algorithms used to classify features. With an accuracy 96.66%, ANN algorithm achieved best classification performance. Afterward, 28 selected with highest gain ratios reclassified. Based features, SVM (Support Vector Machine) 96.66%. expert performed using learning evaluate results obtained through algorithms. When used, was 98.61%, when 97.77%. results, artificial intelligence techniques are suitable for detecting automatically classifying them.

Language: Английский

---

Optimizing Anthracnose Severity Grading in Green Beans with CNN-LSTM Integration

Published: May 24, 2024

Language: Английский

---

Face mask recognition system using MobileNetV2 with optimization function

Applied Artificial Intelligence, Journal Year: 2022, Volume and Issue: 36(1)

Published: Nov. 14, 2022

The world has experienced a health crisis with the outbreak of COVID-19 virus. mask been identified as most effective way to prevent spread This led need for face recognition device that not only detects presence but also provides accuracy which person is wearing mask. In addition, should be recognized from all angles. project aims create new and improved real-time tool using image processing computer vision approaches. A dataset consisting images without was used. For purposes this project, pre-trained MobileNetV2 convolutional neural network performance given model evaluated. presented in can detect an 99.21%. effectively side direction, makes it more useful. optimization function contains learning loops are

Language: Английский

---

Machine Learning-Based Classification of Infected Date Palm Leaves Caused by Dubas Insects: A Comparative Analysis of Feature Extraction Methods and Classification Algorithms

2022 Innovations in Intelligent Systems and Applications Conference (ASYU), Journal Year: 2023, Volume and Issue: unknown

Published: Oct. 11, 2023

This study investigates the utilization of machine learning techniques for effectively classifying infected date palm leaves caused by Dubas insects. Three distinct feature extraction methods, namely Inceptionv3, SqueezeNet, and VGG16, are combined with five diverse algorithms: K-Nearest Neighbors (KNN), Neural Network (ANN), Random Forest (RF), Artificial Support Vector Machine (SVM), Logistic Regression (LR). The dataset comprises a collection images depicting leaves, performance evaluation metrics, including accuracy, recall, precision, F1 score, computed each algorithm. results unveil varied levels accuracy among methods algorithms. Specifically, Inceptionv3 achieved an 80.4% KNN, while SqueezeNet attained 75.3% VGG16 obtained 76.6% accuracy. For SVM, scores were 72.9%, 66%, 62.4%, respectively. ANN demonstrated promising 83.8%, 80%, 80.1% Lastly, LR yielded 83%, 76.2%, 80% These findings offer useful information about how various perform in thereby facilitating development effective pest management strategies plantations.

Language: Английский

---

Detection of Chicken Diseases from Fecal Images with the Pre-Trained Places365-GoogLeNet Model

Published: Sept. 7, 2023

A variety of chicken diseases pose significant challenges to farmers worldwide, posing a threat the safety food and potentially resulting in economic losses. In this study, we propose utilization pre-trained deep learning model, Places365-GoogLeNet, for detection from fecal images, including Healthy, Coccidiosis, Salmonella, New Castle Disease. By leveraging powerful image analysis capabilities learning, our approach achieves remarkable classification accuracy 98.91%. This surpasses results reported related studies literature. Moreover, findings highlight potential artificial intelligence machine techniques, particularly agricultural sector, automated disease detection. The presented not only contribute early diagnosis prompt intervention poultry farming but also pave way future research develop more advanced methods utilize larger diverse datasets enhance model's generalization ability.

Language: Английский

---