Forgery detection algorithm based on texture features

Ahmed, Ismail Taha; Hammad, Baraa Tareq; Jamil, Norziana

doi:10.11591/ijeecs.v24.i1.pp226-235

Cited by 12 publications

(6 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Then, these training sets will be used to train these models. Lastly, the trained KNN model is tested to identify the class of malware that it falls into [41].…”

Section: End For Endmentioning

confidence: 99%

Examining the Performance of Various Pretrained Convolutional Neural Network Models in Malware Detection

Abdulazeez,

Ahmed,

Hammad

2024

Applied Sciences

View full text Add to dashboard Cite

A significant quantity of malware is created on purpose every day. Users of smartphones and computer networks now mostly worry about malware. These days, malware detection is a major concern in the cybersecurity area. Several factors can impact malware detection performance, such as inappropriate features and classifiers, extensive domain knowledge, imbalanced data environments, computational complexity, and resource usage. A significant number of existing malware detection methods have been impacted by these factors. Therefore, in this paper, we will first identify and determine the best features and classifiers and then use them in order to propose the malware detection method. The comparative strategy and proposed malware detection procedure consist of four basic steps: malware transformation (converting images of malware from RGB to grayscale), feature extraction (using the ResNet-50, DenseNet-201, GoogLeNet, AlexNet, and SqueezeNet models), feature selection (using PCA method), classification (including GDA, KNN, logistic, SVM, RF, and ensemble learning), and evaluation (using accuracy and error evaluation metrics). Unbalanced Malimg datasets are used in experiments to validate the efficacy of the results that were obtained. According to the comparison findings, KNN is the best machine learning classifier. It outperformed the other classifiers in the Malimg datasets in terms of both accuracy and error. In addition, DenseNet201 is the best pretrained model in the Malimg dataset. Therefore, the proposed DenseNet201-KNN methods had an accuracy rate of 96% and a minimal error rate of 3.07%. The proposed methods surpass existing state-of-the-art approaches. The proposed feature extraction is computationally quicker than most other methods since it uses a lightweight design and fewer feature vector dimensions.

show abstract

“…Then, these training sets will be used to train these models. Lastly, the trained KNN model is tested to identify the class of malware that it falls into [41].…”

Section: End For Endmentioning

confidence: 99%

Examining the Performance of Various Pretrained Convolutional Neural Network Models in Malware Detection

Abdulazeez,

Ahmed,

Hammad

2024

Applied Sciences

View full text Add to dashboard Cite

show abstract

“…A broad literature review is presented in [21] on image forgery detection with deep learning approach which have accomplished high performance accuracies incontext with image or facial image. Research by Ahmed et al [22] focuses on SFTA, Haralick, and LBP as feature extraction methods followed by features fed to KNN classifier to detect and classify the copy-move attack. The rust shows that k-nearest neighbors (KNN) classifier gives result accuracy of 81.81%, 86.36% and 95.45% for segmentation based fractal texture analysis (SFTA), local binary patterns (LBP) and haralicks respectively.…”

Section: Classification Of Forgery Detectionmentioning

confidence: 99%

A comparative study on image forgery-facial retouching

Sheth

Vora

2023

Bulletin EEI

View full text Add to dashboard Cite

Forgery with the digital images is being very easy now days due to the very advanced and open source image editing tools, software and devises which supports a high quality of resolutions. Tempering with digital documents for changing identity or sometimes for fun is increasing day by day as the era is of digital world. Detecting clues of tampering and verifying the authenticity of images is an important issues now-a-days and growing research field. The existing research in the area of digital image forgery identification is discussed here. Different types of image forgery attacks along with its detection and classification are discussed too. A general review of facial image authentication is studied and discussed here along with the current methods and techniques for the detection and classification. A comparative analysis is presented here for the researchers as a mind mapping to work in the same field.

show abstract

“…Texture analysis is crucial in a variety of applications. Steganalysis is one of the most significant [19], [20]. The reason for this is that the information hidden in the images is very difficult to discover or distinguish with human eye, therefore texture analysis is used to uncover information that the human eye cannot see it [21].…”

Section: Glcm Featurementioning

confidence: 99%

Low feature dimension in image steganographic recognition

Ahmed

Jamil

Hammad

2022

IJEECS

Self Cite

View full text Add to dashboard Cite

Steganalysis <span>aids in the detection of steganographic data without the need to know the embedding algorithm or the "cover" image. The researcher's major goal was to develop a Steganalysis technique that might improve recognition accuracy while utilizing a minimal feature vector dimension. A number of Steganalysis techniques have been developed to detect steganography in images. However, the steganalysis technique's performance is still limited due to their large feature vector dimension, which takes a long time to compute. The variations of texture and properties of an embedded image are clearly seen. Therefore, in this paper, we proposed Steganalysis recognition based on one of the texture features, such as gray level co-occurrence matrix (GLCM). As a classifier, Ada-Boost and Gaussian discriminant analysis (GDA) are used. In order to evaluate the performance of the proposed method, we use a public database in our proposed and applied it using IStego100K datasets. The results of the experiment show that the proposed can improve accuracy greatly. It also indicates that in terms of accuracy, the Ada-Boost classifier surpassed the GDA. The comparative findings show that the proposed method outperforms other current techniques especially in terms of feature size and recognition accuracy</span>.

show abstract

Forgery detection algorithm based on texture features

Cited by 12 publications

References 26 publications

Examining the Performance of Various Pretrained Convolutional Neural Network Models in Malware Detection

Examining the Performance of Various Pretrained Convolutional Neural Network Models in Malware Detection

A comparative study on image forgery-facial retouching

Low feature dimension in image steganographic recognition

Contact Info

Product

Resources

About