Pub Date : 2021-05-01DOI: 10.4018/IJDCF.20210501.OA2
Henan Shi, Tanfeng Sun, Xinghao Jiang, Yi Dong, Ke Xu
The development of video steganography has put forward a higher demand for video steganalysis. This paper presents a novel steganalysis against discrete cosine/sine transform (DCT/DST)-based steganography for high efficiency video coding (HEVC) videos. The new steganalysis employs special frames extraction (SFE) and accordion unfolding (AU) transformation to target the latest DCT/DST domain HEVC video steganography algorithms by merging temporal and spatial correlation. In this article, the distortion process of DCT/DST-based HEVC steganography is firstly analyzed. Then, based on the analysis, two kinds of distortion, the intra-frame distortion and the inter-frame distortion, are mainly caused by DCT/DST-based steganography. Finally, to effectively detect these distortions, an innovative method of HEVC steganalysis is proposed, which gives a combination feature of SFE and a temporal to spatial transformation, AU. The experiment results show that the proposed steganalysis performs better than other methods.
{"title":"A HEVC Video Steganalysis Against DCT/DST-Based Steganography","authors":"Henan Shi, Tanfeng Sun, Xinghao Jiang, Yi Dong, Ke Xu","doi":"10.4018/IJDCF.20210501.OA2","DOIUrl":"https://doi.org/10.4018/IJDCF.20210501.OA2","url":null,"abstract":"The development of video steganography has put forward a higher demand for video steganalysis. This paper presents a novel steganalysis against discrete cosine/sine transform (DCT/DST)-based steganography for high efficiency video coding (HEVC) videos. The new steganalysis employs special frames extraction (SFE) and accordion unfolding (AU) transformation to target the latest DCT/DST domain HEVC video steganography algorithms by merging temporal and spatial correlation. In this article, the distortion process of DCT/DST-based HEVC steganography is firstly analyzed. Then, based on the analysis, two kinds of distortion, the intra-frame distortion and the inter-frame distortion, are mainly caused by DCT/DST-based steganography. Finally, to effectively detect these distortions, an innovative method of HEVC steganalysis is proposed, which gives a combination feature of SFE and a temporal to spatial transformation, AU. The experiment results show that the proposed steganalysis performs better than other methods.","PeriodicalId":44650,"journal":{"name":"International Journal of Digital Crime and Forensics","volume":"26 1","pages":"19-33"},"PeriodicalIF":0.7,"publicationDate":"2021-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90636688","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-05-01DOI: 10.4018/IJDCF.20210501.OA3
Shiqi Wu, Bo Wang, Jianxiang Zhao, Mengnan Zhao, Kun Zhong, Yanqing Guo
Nowadays, source camera identification, which aims to identify the source camera of images, is quite important in the field of forensics. There is a problem that cannot be ignored that the existing methods are unreliable and even out of work in the case of the small training sample. To solve this problem, a virtual sample generation-based method is proposed in this paper, combined with the ensemble learning. In this paper, after constructing sub-sets of LBP features, the authors generate a virtual sample-based on the mega-trend-diffusion (MTD) method, which calculates the diffusion range of samples according to the trend diffusion theory, and then randomly generates virtual sample according to uniform distribution within this range. In the aspect of the classifier, an ensemble learning scheme is proposed to train multiple SVM-based classifiers to improve the accuracy of image source identification. The experimental results demonstrate that the proposed method achieves higher average accuracy than the state-of-the-art, which uses a small number of samples as the training sample set.
{"title":"Virtual Sample Generation and Ensemble Learning Based Image Source Identification With Small Training Samples","authors":"Shiqi Wu, Bo Wang, Jianxiang Zhao, Mengnan Zhao, Kun Zhong, Yanqing Guo","doi":"10.4018/IJDCF.20210501.OA3","DOIUrl":"https://doi.org/10.4018/IJDCF.20210501.OA3","url":null,"abstract":"Nowadays, source camera identification, which aims to identify the source camera of images, is quite important in the field of forensics. There is a problem that cannot be ignored that the existing methods are unreliable and even out of work in the case of the small training sample. To solve this problem, a virtual sample generation-based method is proposed in this paper, combined with the ensemble learning. In this paper, after constructing sub-sets of LBP features, the authors generate a virtual sample-based on the mega-trend-diffusion (MTD) method, which calculates the diffusion range of samples according to the trend diffusion theory, and then randomly generates virtual sample according to uniform distribution within this range. In the aspect of the classifier, an ensemble learning scheme is proposed to train multiple SVM-based classifiers to improve the accuracy of image source identification. The experimental results demonstrate that the proposed method achieves higher average accuracy than the state-of-the-art, which uses a small number of samples as the training sample set.","PeriodicalId":44650,"journal":{"name":"International Journal of Digital Crime and Forensics","volume":"83 1","pages":"34-46"},"PeriodicalIF":0.7,"publicationDate":"2021-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86203893","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-05-01DOI: 10.4018/IJDCF.20210501.OA4
Jianbin Wu, Yang Zhang, Chuwei Luo, L. Yuan, X. Shen
In order to improve the data-embedding capacity of modification-free steganography algorithm, scholars have done a lot of research work to meet practical demands. By researching the user's behavioral habits of several social platforms, a semi-structured modification-free steganography algorithm is introduced in the paper. By constructing the mapping relationship between small icons and binary numbers, the idea of image stitching is utilized, and small icons are stitched together according to the behavioral habits of people's social platforms to implement the graphical representation of secret messages. The convolutional neural network (CNN) has been used to train the small icon recognition and classification data set in the algorithm. In order to improve the robustness of the algorithm, the icons processed by various attack methods are introduced as interference samples in the training set. The experimental results show that the algorithm has good anti-attack ability, and the hiding capacity can be improved, which can be used in the covert communication.
{"title":"A Modification-Free Steganography Algorithm Based on Image Classification and CNN","authors":"Jianbin Wu, Yang Zhang, Chuwei Luo, L. Yuan, X. Shen","doi":"10.4018/IJDCF.20210501.OA4","DOIUrl":"https://doi.org/10.4018/IJDCF.20210501.OA4","url":null,"abstract":"In order to improve the data-embedding capacity of modification-free steganography algorithm, scholars have done a lot of research work to meet practical demands. By researching the user's behavioral habits of several social platforms, a semi-structured modification-free steganography algorithm is introduced in the paper. By constructing the mapping relationship between small icons and binary numbers, the idea of image stitching is utilized, and small icons are stitched together according to the behavioral habits of people's social platforms to implement the graphical representation of secret messages. The convolutional neural network (CNN) has been used to train the small icon recognition and classification data set in the algorithm. In order to improve the robustness of the algorithm, the icons processed by various attack methods are introduced as interference samples in the training set. The experimental results show that the algorithm has good anti-attack ability, and the hiding capacity can be improved, which can be used in the covert communication.","PeriodicalId":44650,"journal":{"name":"International Journal of Digital Crime and Forensics","volume":"1 1","pages":"47-58"},"PeriodicalIF":0.7,"publicationDate":"2021-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90405480","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-03-01DOI: 10.4018/IJDCF.2021030101
Sameena Naaz
Phishing attacks are growing in the similar manner as e-commerce industries are growing. Prediction and prevention of phishing attacks is a very critical step towards safeguarding online transactions. Data mining tools can be applied in this regard as the technique is very easy and can mine millions of information within seconds and deliver accurate results. With the help of machine learning algorithms like random forest, decision tree, neural network, and linear model, we can classify data into phishing, suspicious, and legitimate. The devices that are connected over the internet, known as internet of things (IoT), are also at very high risk of phishing attack. In this work, machine learning algorithms random forest classifier, support vector machine, and logistic regression have been applied on IoT dataset for detection of phishing attacks, and then the results have been compared with previous work carried out on the same dataset as well as on a different dataset. The results of these algorithms have then been compared in terms of accuracy, error rate, precision, and recall.
{"title":"Detection of Phishing in Internet of Things Using Machine Learning Approach","authors":"Sameena Naaz","doi":"10.4018/IJDCF.2021030101","DOIUrl":"https://doi.org/10.4018/IJDCF.2021030101","url":null,"abstract":"Phishing attacks are growing in the similar manner as e-commerce industries are growing. Prediction and prevention of phishing attacks is a very critical step towards safeguarding online transactions. Data mining tools can be applied in this regard as the technique is very easy and can mine millions of information within seconds and deliver accurate results. With the help of machine learning algorithms like random forest, decision tree, neural network, and linear model, we can classify data into phishing, suspicious, and legitimate. The devices that are connected over the internet, known as internet of things (IoT), are also at very high risk of phishing attack. In this work, machine learning algorithms random forest classifier, support vector machine, and logistic regression have been applied on IoT dataset for detection of phishing attacks, and then the results have been compared with previous work carried out on the same dataset as well as on a different dataset. The results of these algorithms have then been compared in terms of accuracy, error rate, precision, and recall.","PeriodicalId":44650,"journal":{"name":"International Journal of Digital Crime and Forensics","volume":"11 1","pages":"1-15"},"PeriodicalIF":0.7,"publicationDate":"2021-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88824644","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-03-01DOI: 10.4018/IJDCF.2021030106
Jin Du, Feng Yuan, Liping Ding, Guangxuan Chen, Xuehua Liu
The study of complex networks is to discover the characteristics of these connections and to discover the nature of the system between them. Link prediction method is a classic in the study of complex networks. It ca not only reflect the relationship between the node similarity. More can be estimated through the edge, which reveals the intrinsic factors of network evolution, namely the network evolution mechanism. Threat information network is the evolution and development of the network. The introduction of such a complex network of interdisciplinary approach is an innovative research perspective to observe that the threat intelligence occurs. The characteristics of the network show, at the same time, also can predict what will happen. The evolution of the network for network security situational awareness of the research provides a new approach.
{"title":"Research on Threat Information Network Based on Link Prediction","authors":"Jin Du, Feng Yuan, Liping Ding, Guangxuan Chen, Xuehua Liu","doi":"10.4018/IJDCF.2021030106","DOIUrl":"https://doi.org/10.4018/IJDCF.2021030106","url":null,"abstract":"The study of complex networks is to discover the characteristics of these connections and to discover the nature of the system between them. Link prediction method is a classic in the study of complex networks. It ca not only reflect the relationship between the node similarity. More can be estimated through the edge, which reveals the intrinsic factors of network evolution, namely the network evolution mechanism. Threat information network is the evolution and development of the network. The introduction of such a complex network of interdisciplinary approach is an innovative research perspective to observe that the threat intelligence occurs. The characteristics of the network show, at the same time, also can predict what will happen. The evolution of the network for network security situational awareness of the research provides a new approach.","PeriodicalId":44650,"journal":{"name":"International Journal of Digital Crime and Forensics","volume":"7 1","pages":"94-102"},"PeriodicalIF":0.7,"publicationDate":"2021-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84650927","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-03-01DOI: 10.4018/IJDCF.2021030105
Partha Ghosh, D. Sarkar, Joy Sharma, S. Phadikar
The present era is being dominated by cloud computing technology which provides services to the users as per demand over the internet. Satisfying the needs of huge people makes the technology prone to activities which come up as a threat. Intrusion detection system (IDS) is an effective method of providing data security to the information stored in the cloud which works by analyzing the network traffic and informs in case of any malicious activities. In order to control high amount of data stored in cloud, data is stored as per relevance leading to distributed computing. To remove redundant data, the authors have implemented data mining process such as feature selection which is used to generate an optimum subset of features from a dataset. In this paper, the proposed IDS provides security working upon the idea of feature selection. The authors have prepared a modified-firefly algorithm which acts as a proficient feature selection method and enables the NSL-KDD dataset to consume less storage space by reducing dimensions as well as less training time with greater classification accuracy.
{"title":"An Intrusion Detection System Using Modified-Firefly Algorithm in Cloud Environment","authors":"Partha Ghosh, D. Sarkar, Joy Sharma, S. Phadikar","doi":"10.4018/IJDCF.2021030105","DOIUrl":"https://doi.org/10.4018/IJDCF.2021030105","url":null,"abstract":"The present era is being dominated by cloud computing technology which provides services to the users as per demand over the internet. Satisfying the needs of huge people makes the technology prone to activities which come up as a threat. Intrusion detection system (IDS) is an effective method of providing data security to the information stored in the cloud which works by analyzing the network traffic and informs in case of any malicious activities. In order to control high amount of data stored in cloud, data is stored as per relevance leading to distributed computing. To remove redundant data, the authors have implemented data mining process such as feature selection which is used to generate an optimum subset of features from a dataset. In this paper, the proposed IDS provides security working upon the idea of feature selection. The authors have prepared a modified-firefly algorithm which acts as a proficient feature selection method and enables the NSL-KDD dataset to consume less storage space by reducing dimensions as well as less training time with greater classification accuracy.","PeriodicalId":44650,"journal":{"name":"International Journal of Digital Crime and Forensics","volume":"95 1","pages":"77-93"},"PeriodicalIF":0.7,"publicationDate":"2021-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78566042","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-03-01DOI: 10.4018/IJDCF.2021030108
Pengcheng Cao, Weiwei Liu, Guangjie Liu, Jiangtao Zhai, Xiaopeng Ji, Yue-wei Dai, Huiwen Bai
To conceal the very existence of communication, the noise-based wireless covert channel modulates secret messages into artificial noise, which is added to the normal wireless signal. Although the state-of-the-art work based on constellation modulation has made the composite and legitimate signal undistinguishable, there exists an imperfection on reliability due to the dense distribution of covert constellation points. In this study, the authors design a wireless covert channel based on dither analog chaotic code to improve the reliability without damaging the undetectability. The dither analog chaotic code (DACC) plays the role as the error correcting code. In the modulation, the analog variables converted from secret messages are encode into joint codewords by chaotic mapping and dither derivation of DACC. The joint codewords are mapped to artificial noise later. Simulation results show that the proposed scheme can achieve better reliability than the state-of-the-art scheme while maintaining the similar performance on undetectability.
{"title":"Design a Wireless Covert Channel Based on Dither Analog Chaotic Code","authors":"Pengcheng Cao, Weiwei Liu, Guangjie Liu, Jiangtao Zhai, Xiaopeng Ji, Yue-wei Dai, Huiwen Bai","doi":"10.4018/IJDCF.2021030108","DOIUrl":"https://doi.org/10.4018/IJDCF.2021030108","url":null,"abstract":"To conceal the very existence of communication, the noise-based wireless covert channel modulates secret messages into artificial noise, which is added to the normal wireless signal. Although the state-of-the-art work based on constellation modulation has made the composite and legitimate signal undistinguishable, there exists an imperfection on reliability due to the dense distribution of covert constellation points. In this study, the authors design a wireless covert channel based on dither analog chaotic code to improve the reliability without damaging the undetectability. The dither analog chaotic code (DACC) plays the role as the error correcting code. In the modulation, the analog variables converted from secret messages are encode into joint codewords by chaotic mapping and dither derivation of DACC. The joint codewords are mapped to artificial noise later. Simulation results show that the proposed scheme can achieve better reliability than the state-of-the-art scheme while maintaining the similar performance on undetectability.","PeriodicalId":44650,"journal":{"name":"International Journal of Digital Crime and Forensics","volume":"36 1","pages":"115-133"},"PeriodicalIF":0.7,"publicationDate":"2021-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86796570","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-03-01DOI: 10.4018/IJDCF.2021030102
Hanlin Liu, Jingju Liu, Xuehu Yan, Pengfei Xue, Dingwei Tan
This paper proposes an audio steganography method based on run length encoding and integer wavelet transform which can be used to hide secret message in digital audio. The major contribution of the proposed scheme is to propose an audio steganography with high capacity, where the secret information is compressed by run length encoding. In the applicable scenario, the main purpose is to hide as more information as possible in the cover audio files. First, the secret information is chaotic scrambling, then the result of scrambling is run length encoded, and finally, the secret information is embedded into integer wavelet coefficients. The experimental results and comparison with existing technique show that by utilizing the lossless compression of run length encoding and anti-attack of wavelet domain, the proposed method has improved the capacity, good audio quality, and can achieve blind extraction while maintaining imperceptibility and strong robustness.
{"title":"An Audio Steganography Based on Run Length Encoding and Integer Wavelet Transform","authors":"Hanlin Liu, Jingju Liu, Xuehu Yan, Pengfei Xue, Dingwei Tan","doi":"10.4018/IJDCF.2021030102","DOIUrl":"https://doi.org/10.4018/IJDCF.2021030102","url":null,"abstract":"This paper proposes an audio steganography method based on run length encoding and integer wavelet transform which can be used to hide secret message in digital audio. The major contribution of the proposed scheme is to propose an audio steganography with high capacity, where the secret information is compressed by run length encoding. In the applicable scenario, the main purpose is to hide as more information as possible in the cover audio files. First, the secret information is chaotic scrambling, then the result of scrambling is run length encoded, and finally, the secret information is embedded into integer wavelet coefficients. The experimental results and comparison with existing technique show that by utilizing the lossless compression of run length encoding and anti-attack of wavelet domain, the proposed method has improved the capacity, good audio quality, and can achieve blind extraction while maintaining imperceptibility and strong robustness.","PeriodicalId":44650,"journal":{"name":"International Journal of Digital Crime and Forensics","volume":"38 1","pages":"16-34"},"PeriodicalIF":0.7,"publicationDate":"2021-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74811818","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-03-01DOI: 10.4018/IJDCF.2021030104
Ning Huang, Shuguang Huang, Chao Chang
W⊕X is a protection mechanism against control-flow hijacking attacks. Return-oriented programming (ROP) can perform a specific function by searching for appropriate assembly instruction fragments (gadgets) in a code segment and bypass the W⊕X. However, manual search for gadgets that match the conditions is inefficient, with high error and missing rates. In order to improve the efficiency of ROP generation, the authors propose an automatic generation method based on a fragmented layout called automatic generation of ROP. This method designs new intermediate instruction construction rules based on an automatic ROP generation framework Q, uses symbolic execution to analyze program memory states and construct data constraints for multi-modules ROP, and solves ROP data constraints to generate test cases of an ROP chain. Experiments show that this method can effectively improve the space efficiency of the ROP chain and lower the requirements of the ROP layout on memory conditions.
{"title":"Automatic Generation of ROP Through Static Instructions Assignment and Dynamic Memory Analysis","authors":"Ning Huang, Shuguang Huang, Chao Chang","doi":"10.4018/IJDCF.2021030104","DOIUrl":"https://doi.org/10.4018/IJDCF.2021030104","url":null,"abstract":"W⊕X is a protection mechanism against control-flow hijacking attacks. Return-oriented programming (ROP) can perform a specific function by searching for appropriate assembly instruction fragments (gadgets) in a code segment and bypass the W⊕X. However, manual search for gadgets that match the conditions is inefficient, with high error and missing rates. In order to improve the efficiency of ROP generation, the authors propose an automatic generation method based on a fragmented layout called automatic generation of ROP. This method designs new intermediate instruction construction rules based on an automatic ROP generation framework Q, uses symbolic execution to analyze program memory states and construct data constraints for multi-modules ROP, and solves ROP data constraints to generate test cases of an ROP chain. Experiments show that this method can effectively improve the space efficiency of the ROP chain and lower the requirements of the ROP layout on memory conditions.","PeriodicalId":44650,"journal":{"name":"International Journal of Digital Crime and Forensics","volume":"35 1","pages":"57-76"},"PeriodicalIF":0.7,"publicationDate":"2021-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"72773766","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-03-01DOI: 10.4018/IJDCF.2021030107
Yongzhen Ke, Yiping Cui
Tampering with images may involve the field of crime and also bring problems such as incorrect values to the public. Image local deformation is one of the most common image tampering methods, where the original texture features and the correlation between the pixels of an image are changed. Multiple fusion strategies based on first-order difference images and their texture feature is proposed to locate the tamper in local deformation image. Firstly, texture features using overlapping blocks on one color channel are extracted and fed into fuzzy c-means clustering method to generate a tamper probability map (TPM), and then several TPMs with different block sizes are fused in the first fusion. Secondly, different TPMs with different color channels and different texture features are respectively fused in the second and third fusion. The experimental results show that the proposed method can accurately detect the location of the local deformation of an image.
{"title":"Multiple Fusion Strategies in Localization of Local Deformation Tampering","authors":"Yongzhen Ke, Yiping Cui","doi":"10.4018/IJDCF.2021030107","DOIUrl":"https://doi.org/10.4018/IJDCF.2021030107","url":null,"abstract":"Tampering with images may involve the field of crime and also bring problems such as incorrect values to the public. Image local deformation is one of the most common image tampering methods, where the original texture features and the correlation between the pixels of an image are changed. Multiple fusion strategies based on first-order difference images and their texture feature is proposed to locate the tamper in local deformation image. Firstly, texture features using overlapping blocks on one color channel are extracted and fed into fuzzy c-means clustering method to generate a tamper probability map (TPM), and then several TPMs with different block sizes are fused in the first fusion. Secondly, different TPMs with different color channels and different texture features are respectively fused in the second and third fusion. The experimental results show that the proposed method can accurately detect the location of the local deformation of an image.","PeriodicalId":44650,"journal":{"name":"International Journal of Digital Crime and Forensics","volume":"46 1","pages":"103-114"},"PeriodicalIF":0.7,"publicationDate":"2021-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77072779","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：