Modern Learning Techniques and Plant Image Classification

Ünal, Metehan; Bostancı, Erkan; Güzel, Mehmet Serdar; Aydın, Ayhan

doi:10.33769/aupse.627897

Cited by 2 publications

(2 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The process is called "multi-hot encoding". As is known, the performance of deep-learning models is generally directly related to the number of data [30]. For this reason, random data augmentation was performed on the training data during the training of the models.…”

Section: The Architecture Of Pretrained Modelsmentioning

confidence: 99%

Multilabel Genre Prediction Using Deep-Learning Frameworks

et al. 2023

View full text Add to dashboard Cite

In this study, transfer learning has been used to overcome multilabel classification tasks. As a case study, movie genre classification by using posters has been chosen. Six state-of-the-art pretrained models, VGG16, ResNet, DenseNet, Inception, MobileNet, and ConvNeXt, have been employed for this experiment. The movie posters have been obtained from Internet Movie Database (IMDB). The dataset has been divided using an iterative stratification technique. A sequence of dense layers has been added on top of each model and these models have been trained and fine-tuned. All the results of the models compared considered accuracy, loss, Hamming loss, F1-score, precision, and AUC metrics. When the metrics used were evaluated, the most successful result regarding accuracy has been obtained from the modified DenseNet architecture at 90%. Also, the ConvNeXt, which is the newest model among all, performed quite satisfactorily, reaching over 90% accuracy. This study uses an iterative stratification method to split an unbalanced dataset which provides more reliable results than the classical splitting method which is the common method in the literature. Also, the feature extraction capabilities of the six pretrained models have been compared. The outcome of this study shows promising results regarding multilabel classification. As for future work, it is planned to enhance this study by using natural language processing and ensemble methods.

show abstract

Section: The Architecture Of Pretrained Modelsmentioning

confidence: 99%

Multilabel Genre Prediction Using Deep-Learning Frameworks

et al. 2023

View full text Add to dashboard Cite

show abstract

“…The work has also presented a range of computer vision techniques and has also provided an illustration of the research in the future. Unal et al published a paper explaining modern learning techniques such as ANN and transfer learning [8]. This paper aims to classify plant seedling images using two CNN architectures to test transfer learning.…”

Section: Literature Reviewmentioning

confidence: 99%

Disease detection in bean leaves using deep learning

SERTTAŞ

Deniz

2023

Communications Faculty of Sciences University of Ankara Series A2-A3 Physical Sciences and Engineering

View full text Add to dashboard Cite

The care and health of agricultural plants, which are the primary source for people to eat healthily, are essential. Disease detection in plants is one of the critical elements of smart agriculture. In parallel with the development of artificial intelligence, advancements in smart agriculture are also progressing. The development of deep learning techniques positively affects smart farming practices. Today, using deep learning and computer vision techniques, various plant diseases can be detected from images such as photographs. In this research, deep learning techniques were used to detect and diagnose bean leaf diseases. Healthy and diseased bean leaf images were used to train the convolutional neural network (CNN) model, which is one of the deep learning techniques. Transfer learning was applied to CNN models to detect plant diseases with the difference of related works. A transfer learning-based strategy to identify various diseases in plant varieties is demonstrated using leaf images of healthy and diseased plants from the Bean dataset. With the proposed method, 1295 images were studied. The results show that our technique successfully identified disease status in bean leaf images, achieving an accuracy of 98.33% with the ResNet50 model.

show abstract

Modern Learning Techniques and Plant Image Classification

Cited by 2 publications

References 13 publications

Multilabel Genre Prediction Using Deep-Learning Frameworks

Multilabel Genre Prediction Using Deep-Learning Frameworks

Disease detection in bean leaves using deep learning

Contact Info

Product

Resources

About