Classification and Analysis of Agaricus bisporus Diseases with Pre-Trained Deep Learning Models

Agronomy, Год журнала: 2025, Номер 15(1), С. 226 - 226

Опубликована: Янв. 17, 2025

This research evaluates 20 advanced convolutional neural network (CNN) architectures for classifying mushroom diseases in Agaricus bisporus, utilizing a custom dataset of 3195 images (2464 infected and 731 healthy mushrooms) captured under uniform white-light conditions. The consistent illumination the enhances robustness practical usability assessed models. Using weighted scoring system that incorporates precision, recall, F1-score, area ROC curve (AUC), average precision (AP), ResNet-50 achieved highest overall score 99.70%, demonstrating outstanding performance across all disease categories. DenseNet-201 DarkNet-53 followed closely, confirming their reliability classification tasks with high recall values. Confusion matrices curves further validated capabilities These findings underscore potential CNN-based approaches accurate efficient early detection diseases, contributing to more sustainable data-driven agricultural practices.

Язык: Английский

---

Convolutional neural network-support vector machine-based approach for identification of wheat hybrids

European Food Research and Technology, Год журнала: 2024, Номер 250(5), С. 1353 - 1362

Опубликована: Фев. 15, 2024

Abstract Selecting wheat hybrids is vital for enhancing crop yield, adapting to changing climates, and ensuring food security. These align with market demands sustainable farming practices, contributing efficient management. Traditional methods hybrid selection, such as molecular techniques, are costly time-consuming, prone human error. However, advancements in artificial intelligence machine learning offer non-destructive, objective, more solutions. This study explored the classification of varieties using two deep models, MobileNetv2 GoogleNet. models achieved impressive accuracy, reaching 99.26% GoogleNet achieving 97.41%. In second scenario, features obtained from these classified Support Vector Machine (SVM). made MobileNetv2-SVM model, an accuracy 99.91% achieved. provided rapid accurate variety identification method, well breeding programs

Язык: Английский

---

Analysis of selected deep features with CNN-SVM-based for bread wheat seed classification

European Food Research and Technology, Год журнала: 2024, Номер 250(6), С. 1551 - 1561

Опубликована: Март 14, 2024

Abstract The main ingredient of flour is processed wheat. Wheat an agricultural product that harvested once a year. It may be necessary to choose the variety wheat for growing and efficient harvesting. important its economic value, taste, crop yield. Although there are many varieties wheat, they very similar in colour, size, shape, it requires expertise distinguish them by eye. This time consuming can lead human error. Using computer vision artificial intelligence, such problems solved more quickly objectively. In this study, attempt was made classify five bread belonging different cultivars using Convolutional Neural Network (CNN) models. Three approaches have been proposed classification. First, pre-trained CNN models (ResNet18, ResNet50, ResNet101) were trained cultivars. Second, features extracted from fc1000 layer ResNet18, ResNet101 classified support vector machine (SVM) classifier with kernel learning techniques classification variants. Finally, SVM methods used second stage obtained optimal set represent all minimum redundancy maximum relevance (mRMR) feature selection algorithm.The accuracies first, second, last phases as follows. first phase, most successful method classifying grains ResNet18 model 97.57%. + ResNet50 Quadratic ResNet 94.08%.The accuracy 1000 effective selected algorithm 94.51%. slightly lower than deep learning, much shorter 93%. result confirms great effectiveness grain

Язык: Английский

---

Ensemble and optimization algorithm in support vector machines for classification of wheat genotypes

Scientific Reports, Год журнала: 2024, Номер 14(1)

Опубликована: Сен. 30, 2024

Язык: Английский

---

Webserver-Based Mobile Application for Multi-class Chestnut (Castanea sativa) Classification Using Deep Features and Attention Mechanisms

Deleted Journal, Год журнала: 2025, Номер 67(3)

Опубликована: Май 5, 2025

Язык: Английский

---

Harnessing deep learning for wheat variety classification: a convolutional neural network and transfer learning approach

Journal of the Science of Food and Agriculture, Год журнала: 2025, Номер unknown

Опубликована: Май 24, 2025

Abstract BACKGROUND Computer vision and the use of image‐based solutions are gaining traction as non‐destructive food assessment methods because low costs computational equipment. Research conducted on development wheat classification models has been based limited data a smaller number classes compared to availability varieties. To assess applicability convolutional neural network (CNN) models, present study prepared multi‐view images 124 Using deep learning (DL) methods, four‐layered CNN model was developed from scratch, popular architectures, DenseNet201, MobileNet InceptionV3 were trained using transfer learning. RESULTS The proposed model, achieved accuracies 95.40%, 92.41%, 90.54% 83.47%, respectively, they found be both promising successful. Despite challenges related high resource demands, newly outperformed pretrained models. It can inferred that multi‐view, large‐image dataset contributed significantly model's success in achieving accuracy challenging task classifying CONCLUSION recommends further fine‐tuning hyperparameters improve identify better configurations. Besides, other should evaluated. Moreover, by freezing specific early layers, performed maximize accuracy. Additionally, image datasets used will publicly available allow researchers discover new methodologies classify © 2025 Author(s). Journal Science Food Agriculture published John Wiley & Sons Ltd behalf Society Chemical Industry.

Язык: Английский

---

Agaricus bisporus’ta Görüntü Tabanlı Hastalık Sınıflandırması için Kapsamlı Veri Seti

Mantar dergisi, Год журнала: 2024, Номер 15(1), С. 29 - 42

Опубликована: Апрель 30, 2024

Bu makale, Agaricus bisporus (J.E. Lange) Imbach’un kültüründe görülen hastalıkların sınıflandırması için görüntü tabanlı bir veri seti oluşturulması ve analiz edilmesi üzerine yapılan araştırmayı ele almaktadır. Veri seti, sağlıklı farklı hastalık sınıflarına ait görüntüleri içermektedir. Farklı aydınlatma koşullarında elde edilen görüntüler, ayrı sınıflandırma problemi kullanılabilecek uygunlukta veriler sunmaktadır. araştırma, mantar hastalıklarının tanımlanması sınıflandırılması setinin oluşturulması, otomatik olarak sınıflandırılmasını mümkün kılacak derin öğrenme veya diğer makine öğrenmesi tekniklerinin kullanılmasına imkân sağlayacaktır. sürecinde, çalışma kapsamında geliştirilmiş olan taşınabilir görüntüleme sistemi ile işletmeleri ziyaretleri gerçekleştirilmiş; yaklaşık 7250 adet hastalıklı mantar, 1800 de görüntüsü edilmiştir (Her ortamı 3000 adet). Kültür mantarlarında yaygın 4 sınıf gözlemlenmiştir. Her 3 ortamında görüntülenmiştir.

---

Identifying defects and varieties of Malting Barley Kernels

Scientific Reports, Год журнала: 2024, Номер 14(1)

Опубликована: Сен. 27, 2024

This study introduces a comprehensive approach for classifying individual malting barley kernels, involving dual-sided kernel imaging, specifically designed image processing algorithm, an optimized deep neural network architecture, and mechanical sorting system. The proposed method achieves precise classification into multiple classes, aligning with quality standards material assessment. Throughout the study, various analysis techniques were assessed, including traditional feature engineering, established transfer learning architectures, our custom-designed convolutional tailored analysis. Comparative underscores superior performance of model. reveals that 94% accuracy in defects varieties, outperforming well-established models to complex architectures attain 93% accuracy. Additionally, it surpasses machine extraction support vector classifiers, which achieve below 90% detecting defective kernels 70% varietal classification. However, we also noted approach's advantage morphological recognition. observation guides new research toward integrating modern networks. paper presents cereal images two applications: defect variety It emphasizes importance standardizing orientation merging from both sides kernel, device acquisition fulfills this need.

Язык: Английский

---

Hybridizing Long Short-Term Memory and Bi-Directional Long Short-Term Memory Models for Efficient Classification: A Study on Xanthomonas axonopodis pv. phaseoli (XaP) in Two Bean Varieties

Agronomy, Год журнала: 2024, Номер 14(7), С. 1495 - 1495

Опубликована: Июль 10, 2024

This study was conducted on Xanthomonas axonopodis pv, which causes significant economic losses in the agricultural sector. Here, we a common bacterial blight disease caused by phaseoli (XaP) pathogen Üstün42 and Akbulut bean genera. In this study, total of 4000 images, healthy diseased, were used for both breeds. These images classified AlexNet, VGG16, VGG19 models. Later, reclassification performed applying pre-processing to raw images. According results obtained, accuracy rates pre-processed VGG19, VGG16 AlexNet models determined as 0.9213, 0.9125 0.8950, respectively. The then hybridized with LSTM BiLSTM new created. When performance these hybrid evaluated, it found that more successful than simple models, while gave better LSTM. particular, VGG19+BiLSTM model attracted attention achieving 94.25% classification emphasizes effectiveness image processing techniques agriculture field detection is important dataset literature evaluating

Язык: Английский

---

Machine Learning-Based Classification of Mulberry Leaf Diseases

Proceedings of international conference on intelligent systems and new applications., Год журнала: 2024, Номер unknown

Опубликована: Апрель 28, 2024

This research examines the potential of machine learning methods in classification Mulberry leaf diseases. By applying SqueezeNet's deep feature extraction, study aimed to identify disease patterns efficiently. The dataset used consisted ten distinct classes diseases, which was divided into an 80% training set and a 20% testing set. Support Vector Machine (SVM) supervised algorithm classify model achieved accuracy 77.5%. results demonstrate effectiveness approaches aiding detection management can contribute advancements agricultural monitoring mitigation strategies.

Язык: Английский

---