Kurumun Atıf Alan Makalesi
Atıf Alan Yayın
A CNN-SVM study based on selected deep features for grapevine leaves classification
Scopus
Toplam 166 atıf DOI
The main product of grapevines is grapes that are consumed fresh or processed. In addition, grapevine leaves are harvested once a year as a by-product. The species of grapevine leaves are important in terms of price and taste. In this study, deep learning-based classification is conducted by using images of grapevine leaves. For this purpose, images of 500 vine leaves belonging to 5 species were taken with a special self-illuminating system. Later, this number was increased to 2500 with data augmentation methods. The classification was conducted with a state-of-art CNN model fine-tuned MobileNetv2. As the second approach, features were extracted from pre-trained MobileNetv2′s Logits layer and classification was made using various SVM kernels. As the third approach, 1000 features extracted from MobileNetv2′s Logits layer were selected by the Chi-Squares method and reduced to 250. Then, classification was made with various SVM kernels using the selected features. The most successful method was obtained by extracting features from the Logits layer and reducing the feature with the Chi-Squares method. The most successful SVM kernel was Cubic. The classification success of the system has been determined as 97.60%. It was observed that feature selection increased the classification success although the number of features used in classification decreased.
Atıf Kaynağı
Atıf Yapan Yayın
Hybrid Feature-Based Two-Stage Framework for Audio Deepfake Detection and Generative Model Attribution
Scopus
Havuzumuzda
Rapid advances in technology have made it difficult to distinguish fake voices from natural human voices, bringing serious security risks such as fraud and manipulation. While most studies in the current literature focus solely on binary classifications that question the authenticity of the voice, tracking studies that identify the source of the fake voice and issues of data imbalance are often overlooked. Since accurate source attribution is critical for forensic investigation and preventing misuse, this study proposes a two-stage hierarchical detection system to address both detection and attribution challenges. The system is evaluated on a comprehensive dataset containing real voices and fake voices generated by eight different models, including FlashSpeech, VALLE, and OpenAI. To tackle class imbalance and improve generalization, the study utilizes a 2189-dimensional hybrid feature vector derived from models like YAMNet, Wav2Vec 2.0, and Resemblyzer, combined with 17 different data augmentation techniques that enhance the model's representational power. In the comparative experiments involving nine machine learning algorithms, the SVM (RBF Kernel) model achieved the highest accuracy of 97.97% in the first stage, while Logistic Regression demonstrated the best performance in the second stage with 93.98% accuracy. Considering the averaged accuracy across both stages, SVM (RBF Kernel) exhibited the strongest overall performance with 95.51%. These results demonstrate that the proposed hierarchical approach, coupled with hybrid feature extraction and data augmentation, offers an effective solution for cybersecurity and digital forensics, particularly in preventing the misuse of synthetic voice generation and in digital content verification processes.
Atıf Yapan Makale Bilgileri
Kurumlar (1)
Selçuk Üniversitesi
Selçuklu, Turkey