CNN-Based Ternary Classification for Image Steganalysis

Kang, Sanghoon; Park, Hanhoon; Park, Jong-Il

doi:10.3390/electronics8111225

Cited by 6 publications

(5 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus, the network One of the reasons that DL is popular now is the existence of the Convolutional Neural Network (CNN). This CNN has been employed to solve several computer vision tasks [19,20]. It has shown great performance in different medical applications [21,22].…”

Section: Deep Learning (Dl)mentioning

confidence: 99%

Optimizing the Performance of Breast Cancer Classification by Employing the Same Domain Transfer Learning from Hybrid Deep Convolutional Neural Network Model

et al. 2020

View full text Add to dashboard Cite

Breast cancer is a significant factor in female mortality. An early cancer diagnosis leads to a reduction in the breast cancer death rate. With the help of a computer-aided diagnosis system, the efficiency increased, and the cost was reduced for the cancer diagnosis. Traditional breast cancer classification techniques are based on handcrafted features techniques, and their performance relies upon the chosen features. They also are very sensitive to different sizes and complex shapes. However, histopathological breast cancer images are very complex in shape. Currently, deep learning models have become an alternative solution for diagnosis, and have overcome the drawbacks of classical classification techniques. Although deep learning has performed well in various tasks of computer vision and pattern recognition, it still has some challenges. One of the main challenges is the lack of training data. To address this challenge and optimize the performance, we have utilized a transfer learning technique which is where the deep learning models train on a task, and then fine-tune the models for another task. We have employed transfer learning in two ways: Training our proposed model first on the same domain dataset, then on the target dataset, and training our model on a different domain dataset, then on the target dataset. We have empirically proven that the same domain transfer learning optimized the performance. Our hybrid model of parallel convolutional layers and residual links is utilized to classify hematoxylin–eosin-stained breast biopsy images into four classes: invasive carcinoma, in-situ carcinoma, benign tumor and normal tissue. To reduce the effect of overfitting, we have augmented the images with different image processing techniques. The proposed model achieved state-of-the-art performance, and it outperformed the latest methods by achieving a patch-wise classification accuracy of 90.5%, and an image-wise classification accuracy of 97.4% on the validation set. Moreover, we have achieved an image-wise classification accuracy of 96.1% on the test set of the microscopy ICIAR-2018 dataset.

show abstract

Section: Deep Learning (Dl)mentioning

confidence: 99%

Optimizing the Performance of Breast Cancer Classification by Employing the Same Domain Transfer Learning from Hybrid Deep Convolutional Neural Network Model

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Thus, we conducted an experiment using a CNN that is a variant of XuNet [15] (refer to Fig. 1 (a)), which has been previously proposed for ternary classification in [9], where the output of the fully-connected layer is extended to 5. As shown in Table 1, the quinary classification accuracy was very low and the PVD stego and S-UNIWARD stego images were not classified (over-fitting occurred and this is why the classification accuracy for WOW stego images is excessively high).…”

Section: Using a Plain Cnn Or Resnetmentioning

confidence: 99%

“…Unlike conventional CNN-based methods, the proposed method is designed for multi-class classification. A similar method has been proposed for ternary classification of cover, WOW stego and S-UNIWARD stego images using a plain CNN [9]. However, as the number of classes increases, the classification accuracy of this previous method significantly decreases.…”

Section: Introductionmentioning

confidence: 99%

Identification of Multiple Image Steganographic Methods Using Hierarchical ResNets

Kang

Park

2021

IEICE Trans. Inf. & Syst.

Self Cite

View full text Add to dashboard Cite

Image deformations caused by different steganographic methods are typically extremely small and highly similar, which makes their detection and identification to be a difficult task. Although recent steganalytic methods using deep learning have achieved high accuracy, they have been made to detect stego images to which specific steganographic methods have been applied. In this letter, a staganalytic method is proposed that uses hierarchical residual neural networks (ResNet), allowing detection (i.e. classification between stego and cover images) and identification of four spatial steganographic methods (i.e. LSB, PVD, WOW and S-UNIWARD). Experimental results show that using hierarchical ResNets achieves a classification rate of 79.71% in quinary classification, which is approximately 23% higher compared to using a plain convolutional neural network (CNN).

show abstract

“…Convolutional Neural Networks (CNNs) have demonstrated exceptional performance in computer vision and image processing competitions [14]. For instance, [15] achieved precise classification of diabetic foot ulcers using a modified CNN, and [16] achieved ternary classification. However, the electrospun fiber image dataset used in this study is sparse, which often results in poor training results due to the limited data.…”

Section: Introductionmentioning

confidence: 99%

Automatic classification of electrospun fiber polarized light images using mueller matrix depolarization parameter

Guan,

Zou

2023

Front. Phys.

View full text Add to dashboard Cite

Electrospun fibers are widely used in various fields of biology, medicine, and chemistry due to their unique morphological characteristics that determine their distinct application properties. Accurate and rapid classification of these fibers based on their morphology is critical for their effective utilization. Non-destructive and low-cost imaging methods are highly desirable for this purpose, so we obtained the polarization images of different forms of electrospun fibers (smooth surfaces, microporous, and beaded microspheres) by polarized light microscopy. In this study, we have explored the automatic classification of electrospun fibers based on their Mueller matrix depolarization parameter, which is highly correlated with the rough microporous structures on the surface of the object. To achieve this, we employed transfer learning and various convolutional neural networks (CNNs). Our proposed method outperformed the conventional approach that only utilizes a single Mueller matrix M44 image for classification, thus enabling researchers to effectively classify electrospun fibers. Given the high accuracy of our method, it may find significant utility in fields such as material science, nanotechnology, and bioengineering.

show abstract

CNN-Based Ternary Classification for Image Steganalysis

Cited by 6 publications

References 28 publications

Optimizing the Performance of Breast Cancer Classification by Employing the Same Domain Transfer Learning from Hybrid Deep Convolutional Neural Network Model

Optimizing the Performance of Breast Cancer Classification by Employing the Same Domain Transfer Learning from Hybrid Deep Convolutional Neural Network Model

Identification of Multiple Image Steganographic Methods Using Hierarchical ResNets

Automatic classification of electrospun fiber polarized light images using mueller matrix depolarization parameter

Contact Info

Product

Resources

About