Character extraction in the video is very helpful to the understanding of the video content, especially the artificially superimposed characters such as time and place in the surveillance video. However, the performance of the existing algorithms does not meet the needs of application. Therefore, the authors improve semisupervised surveillance video character extraction and recognition with attentional learning multiframe feature fusion. First, the multiframe fusion strategy based on an attention mechanism is adopted to solve the target missing problem, and the Dense ASPP network is introduced to solve the character multiscale problem. Second, a character image denoising algorithm based on semisupervised fuzzy C-means clustering is proposed to isolate and extract clean binary character images. Finally, for some video characters that may involve privacy, traditional and deep learning-based video restoration algorithms are used for characteristic elimination.
{"title":"Semisupervised Surveillance Video Character Extraction and Recognition With Attentional Learning Multiframe Fusion","authors":"Guiyan Cai, Liang Qu, Yongdong Li, Guoan Cheng, Xin Lu, Yiqi Wang, Fengqin Yao, Shengke Wang","doi":"10.4018/ijdcf.315745","DOIUrl":"https://doi.org/10.4018/ijdcf.315745","url":null,"abstract":"Character extraction in the video is very helpful to the understanding of the video content, especially the artificially superimposed characters such as time and place in the surveillance video. However, the performance of the existing algorithms does not meet the needs of application. Therefore, the authors improve semisupervised surveillance video character extraction and recognition with attentional learning multiframe feature fusion. First, the multiframe fusion strategy based on an attention mechanism is adopted to solve the target missing problem, and the Dense ASPP network is introduced to solve the character multiscale problem. Second, a character image denoising algorithm based on semisupervised fuzzy C-means clustering is proposed to isolate and extract clean binary character images. Finally, for some video characters that may involve privacy, traditional and deep learning-based video restoration algorithms are used for characteristic elimination.","PeriodicalId":44650,"journal":{"name":"International Journal of Digital Crime and Forensics","volume":"10 1","pages":"1-15"},"PeriodicalIF":0.7,"publicationDate":"2022-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74265008","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}
File systems are important sources of intelligence information and digital evidence. They have long attracted the interest of researchers in recovering files that are deleted from a hard disk. Existing file recovery studies rely heavily on an operating system (OS). However, it is often encountered that OS services are not available, making existing file recovery approaches unusable. To address this issue, the authors design and implement an OS-independent file recovery algorithm named Golden Eye (GE) by targeting the EXT4 file system. Fed the raw image obtained from a (sanitized) hard disk, GE can automatically recover any designated file or even the whole EXT4 file system. GE is based on the understanding of the file disk layout of EXT4 and does not need any support from additional hardware or software. Experimental results prove the feasibility and correctness of GE. This work not only solves the OS dependency problem that most existing file recovery work suffers from but also reveals the fact that even sanitized hard disks are still at risk of leaking sensitive data.
{"title":"Golden Eye: An OS-Independent Algorithm for Recovering Files From Hard-Disk Raw Images","authors":"Fan Zhang, Wei Chen, Yongqiong Zhu","doi":"10.4018/ijdcf.315793","DOIUrl":"https://doi.org/10.4018/ijdcf.315793","url":null,"abstract":"File systems are important sources of intelligence information and digital evidence. They have long attracted the interest of researchers in recovering files that are deleted from a hard disk. Existing file recovery studies rely heavily on an operating system (OS). However, it is often encountered that OS services are not available, making existing file recovery approaches unusable. To address this issue, the authors design and implement an OS-independent file recovery algorithm named Golden Eye (GE) by targeting the EXT4 file system. Fed the raw image obtained from a (sanitized) hard disk, GE can automatically recover any designated file or even the whole EXT4 file system. GE is based on the understanding of the file disk layout of EXT4 and does not need any support from additional hardware or software. Experimental results prove the feasibility and correctness of GE. This work not only solves the OS dependency problem that most existing file recovery work suffers from but also reveals the fact that even sanitized hard disks are still at risk of leaking sensitive data.","PeriodicalId":44650,"journal":{"name":"International Journal of Digital Crime and Forensics","volume":"40 1","pages":"1-23"},"PeriodicalIF":0.7,"publicationDate":"2022-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86325408","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}
Shannan Liu, Ronghua Zhang, Changzheng Liu, Chenxi Xu, Jie Zhou, Jiaojiao Wang
An improved practical Byzantine fault tolerance (Practical Byzantine Fault Tolerant consensus algorithm based on reputation, RPBFT) algorithm based on grouping and reputation value voting is proposed for the problems of high communication complexity, poor scalability, and random selection of master nodes of the practical Byzantine fault tolerance (PBFT) consensus algorithm of the consortium chain. First, the consistency process is optimized to take the response speed of nodes to each group leader as the basis of grouping, and the intragroup consensus is performed. The group leader then takes the result of intragroup consensus and participates in extra-group consensus to reduce the frequency and time of internode communication. Second, the reputation model and voting mechanism are proposed, and the group leader is generated by node reputation value voting, which enhances the initiative and reliability of trusted nodes and reduces the abnormal nodes as group leader. Finally, a simulation and performance testing system based on this improved scheme is built to prove the effectiveness as well as the usability of the scheme through simulation experiments. The experimental results show that when the number of network nodes is 36, the throughput of the RPBFT algorithm is six times that of PBFT. Therefore, the consensus delay is reduced by 91.7%, and the communication overhead is reduced by 37.8%.
针对联盟链实用拜占庭容错(practical Byzantine fault tolerance, PBFT)共识算法通信复杂度高、可扩展性差、主节点选择随机等问题，提出了一种基于分组和声誉值投票的改进实用拜占庭容错(practical Byzantine fault tolerance consensus algorithm based on reputation, RPBFT)算法。首先，对一致性过程进行优化，以节点对各组长的响应速度作为分组的依据，并进行组内共识。然后，组长接受组内共识的结果，参与组外共识，以减少节点间沟通的频率和时间。其次，提出了信誉模型和投票机制，通过节点信誉值投票产生组长，增强了可信节点的主动性和可靠性，减少了作为组长的异常节点;最后，基于该改进方案搭建了仿真与性能测试系统，通过仿真实验验证了该方案的有效性和可用性。实验结果表明，当网络节点数为36时，RPBFT算法的吞吐量是PBFT算法的6倍。因此，共识延迟减少了91.7%，通信开销减少了37.8%。
{"title":"Improvement of the PBFT Algorithm Based on Grouping and Reputation Value Voting","authors":"Shannan Liu, Ronghua Zhang, Changzheng Liu, Chenxi Xu, Jie Zhou, Jiaojiao Wang","doi":"10.4018/ijdcf.315615","DOIUrl":"https://doi.org/10.4018/ijdcf.315615","url":null,"abstract":"An improved practical Byzantine fault tolerance (Practical Byzantine Fault Tolerant consensus algorithm based on reputation, RPBFT) algorithm based on grouping and reputation value voting is proposed for the problems of high communication complexity, poor scalability, and random selection of master nodes of the practical Byzantine fault tolerance (PBFT) consensus algorithm of the consortium chain. First, the consistency process is optimized to take the response speed of nodes to each group leader as the basis of grouping, and the intragroup consensus is performed. The group leader then takes the result of intragroup consensus and participates in extra-group consensus to reduce the frequency and time of internode communication. Second, the reputation model and voting mechanism are proposed, and the group leader is generated by node reputation value voting, which enhances the initiative and reliability of trusted nodes and reduces the abnormal nodes as group leader. Finally, a simulation and performance testing system based on this improved scheme is built to prove the effectiveness as well as the usability of the scheme through simulation experiments. The experimental results show that when the number of network nodes is 36, the throughput of the RPBFT algorithm is six times that of PBFT. Therefore, the consensus delay is reduced by 91.7%, and the communication overhead is reduced by 37.8%.","PeriodicalId":44650,"journal":{"name":"International Journal of Digital Crime and Forensics","volume":"17 1","pages":"1-15"},"PeriodicalIF":0.7,"publicationDate":"2022-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79645674","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}
Currently, with the popularity of the internet, people are surrounded by a large number of unhealthy pages which have a serious impact on the physical and mental health of visitors. To protect the legitimate rights and interests of internet users from infringement and maintain the harmonious and stable development of society, a new unhealthy webpage discovery system is needed. First, this paper proposed the knowledge of unhealthy webpages and web crawlers, and then the whole system's plan and design were introduced. The test results show that the unhealthy webpage discovery system can meet the needs of users. This experiment uses a CNN algorithm to classify the text and completes the collection and classification of unhealthy information through URL acquisition and URL filtering. The experimental results show that the unhealthy webpage discovery system based on a convolutional neural network can greatly improve the accuracy of unhealthy webpage discovery and reduce the omission rate, which can meet the needs of users for unhealthy webpage discovery.
{"title":"An Unhealthy Webpage Discovery System Based on Convolutional Neural Network","authors":"Zengyu Cai, Chunchen Tan, Jianwei Zhang, Tengteng Xiao, Yuan Feng","doi":"10.4018/ijdcf.315614","DOIUrl":"https://doi.org/10.4018/ijdcf.315614","url":null,"abstract":"Currently, with the popularity of the internet, people are surrounded by a large number of unhealthy pages which have a serious impact on the physical and mental health of visitors. To protect the legitimate rights and interests of internet users from infringement and maintain the harmonious and stable development of society, a new unhealthy webpage discovery system is needed. First, this paper proposed the knowledge of unhealthy webpages and web crawlers, and then the whole system's plan and design were introduced. The test results show that the unhealthy webpage discovery system can meet the needs of users. This experiment uses a CNN algorithm to classify the text and completes the collection and classification of unhealthy information through URL acquisition and URL filtering. The experimental results show that the unhealthy webpage discovery system based on a convolutional neural network can greatly improve the accuracy of unhealthy webpage discovery and reduce the omission rate, which can meet the needs of users for unhealthy webpage discovery.","PeriodicalId":44650,"journal":{"name":"International Journal of Digital Crime and Forensics","volume":"107 1","pages":"1-15"},"PeriodicalIF":0.7,"publicationDate":"2022-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75773379","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}
D. Zang, Xinlei Sheng, Liya Wang, Aimin Yang, Tao Xue, Jie Li
Multitarget threat evaluation of warship air attacks is one of the most urgent problems in warship defense operations. To evaluate the target threat quickly and accurately, an air attack multitarget threat evaluation method based on improved TOPSIS gray relational analysis is proposed. This method establishes threat assessment system of five attributes of target type, anti-jamming ability, heading angle, altitude, and speed. The weight coefficient of each index of the warship is obtained by combining the entropy weight method with the analytic hierarchy process. Topsis can make full use of the information of the original data, and its results can accurately reflect the gap between various evaluation schemes. The weighted Mahalanobis distance and comprehensive gray correlation between the attribute to be evaluated and the positive and negative ideal states are calculated by the improved TOPSIS gray correlation method. The target threat degree to be evaluated is obtained by combining the two methods. Finally, an example is given to prove the effectiveness of the evaluation model.
{"title":"Research and Application of Warship Multiattribute Threat Assessment Based on Improved TOPSIS Gray Association Analysis","authors":"D. Zang, Xinlei Sheng, Liya Wang, Aimin Yang, Tao Xue, Jie Li","doi":"10.4018/ijdcf.315288","DOIUrl":"https://doi.org/10.4018/ijdcf.315288","url":null,"abstract":"Multitarget threat evaluation of warship air attacks is one of the most urgent problems in warship defense operations. To evaluate the target threat quickly and accurately, an air attack multitarget threat evaluation method based on improved TOPSIS gray relational analysis is proposed. This method establishes threat assessment system of five attributes of target type, anti-jamming ability, heading angle, altitude, and speed. The weight coefficient of each index of the warship is obtained by combining the entropy weight method with the analytic hierarchy process. Topsis can make full use of the information of the original data, and its results can accurately reflect the gap between various evaluation schemes. The weighted Mahalanobis distance and comprehensive gray correlation between the attribute to be evaluated and the positive and negative ideal states are calculated by the improved TOPSIS gray correlation method. The target threat degree to be evaluated is obtained by combining the two methods. Finally, an example is given to prove the effectiveness of the evaluation model.","PeriodicalId":44650,"journal":{"name":"International Journal of Digital Crime and Forensics","volume":"1 1","pages":"1-14"},"PeriodicalIF":0.7,"publicationDate":"2022-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90196825","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}
W. Jiang, Ce Zhang, Di Liu, Kaiwei Liu, Zhichao Sun, Jianyuan Wang, Zhongyin Qiu, W. Lv
Aiming at the current software cost model and optimal release research, which does not fully consider the actual faults in the testing phase, a cost-reliability SRGM evaluation and selection algorithm SESABCRC is proposed. From the perspective of incomplete debugging, introducing new faults, and considering testing effort, the imperfect debugging SRGM is established. The proposed SRGM can be used to describe the testing process of the software through the actual failure data set verification, and is superior to other models. Based on the proposed SRGM, the corresponding cost function is given, which explicitly considers the impact of imperfect debugging on the cost. Furthermore, an optimal release strategy is proposed when given restricted reliability target requirements and when considering the uncertainty that the actual cost may exceed the expected cost. Finally, an experimental example is given to illustrate and verify the optimal publishing problem, and parameter sensitivity analysis is carried out.
{"title":"SRGM Decision Model Considering Cost-Reliability","authors":"W. Jiang, Ce Zhang, Di Liu, Kaiwei Liu, Zhichao Sun, Jianyuan Wang, Zhongyin Qiu, W. Lv","doi":"10.4018/ijdcf.302873","DOIUrl":"https://doi.org/10.4018/ijdcf.302873","url":null,"abstract":"Aiming at the current software cost model and optimal release research, which does not fully consider the actual faults in the testing phase, a cost-reliability SRGM evaluation and selection algorithm SESABCRC is proposed. From the perspective of incomplete debugging, introducing new faults, and considering testing effort, the imperfect debugging SRGM is established. The proposed SRGM can be used to describe the testing process of the software through the actual failure data set verification, and is superior to other models. Based on the proposed SRGM, the corresponding cost function is given, which explicitly considers the impact of imperfect debugging on the cost. Furthermore, an optimal release strategy is proposed when given restricted reliability target requirements and when considering the uncertainty that the actual cost may exceed the expected cost. Finally, an experimental example is given to illustrate and verify the optimal publishing problem, and parameter sensitivity analysis is carried out.","PeriodicalId":44650,"journal":{"name":"International Journal of Digital Crime and Forensics","volume":"34 1","pages":"1-19"},"PeriodicalIF":0.7,"publicationDate":"2022-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85817015","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}
The Industrial Control System (ICS) has become the key concept in the modern industrial world, enabling process monitoring and system control for general industrial systems and critical infrastructures. High-skilled hackers can invade an imperfect ICS by existing vulnerabilities without much effort. Conventional defenses (such as encryption and firewall) to keep invaders away are getting less and less effective when an attack is carried out by exploiting an array of particular vulnerabilities. Under this circumstance, a new-type intrusion detection system (IDS) based on deception strategy using honeypot technique is proposed, which is of dramatic effectiveness in protecting ICSs of critical infrastructures. In this honeypot-based model, we capture malicious Internet flows and system operations. We analyze the collected data before alerting and preventing the intrusion alike when it affects the system in the future. This paper deals with the model's concept, architecture, deployment, and what else can be achieved in the field of Critical Infrastructure Cybersecurity (CIC).
{"title":"A Novel IDS Securing Industrial Control System of Critical Infrastructure Using Deception Technology","authors":"Shaobo Zhang, Yuhang Liu, Dequan Yang","doi":"10.4018/ijdcf.302874","DOIUrl":"https://doi.org/10.4018/ijdcf.302874","url":null,"abstract":"The Industrial Control System (ICS) has become the key concept in the modern industrial world, enabling process monitoring and system control for general industrial systems and critical infrastructures. High-skilled hackers can invade an imperfect ICS by existing vulnerabilities without much effort. Conventional defenses (such as encryption and firewall) to keep invaders away are getting less and less effective when an attack is carried out by exploiting an array of particular vulnerabilities. Under this circumstance, a new-type intrusion detection system (IDS) based on deception strategy using honeypot technique is proposed, which is of dramatic effectiveness in protecting ICSs of critical infrastructures. In this honeypot-based model, we capture malicious Internet flows and system operations. We analyze the collected data before alerting and preventing the intrusion alike when it affects the system in the future. This paper deals with the model's concept, architecture, deployment, and what else can be achieved in the field of Critical Infrastructure Cybersecurity (CIC).","PeriodicalId":44650,"journal":{"name":"International Journal of Digital Crime and Forensics","volume":"1 1","pages":"1-20"},"PeriodicalIF":0.7,"publicationDate":"2022-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83863129","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}
In the information age, hackers will use Web vulnerabilities to infiltrate websites, resulting in many security incidents. To solve this problem, security-conscious enterprises or individuals will conduct penetration tests on websites to test and analyze the security of websites, but penetration tests often take a lot of time. Therefore, based on the traditional Web vulnerability scanner, the Web vulnerability detection analyzer designed in this article uses vulnerability detection technologies such as sub-domain scanning, application fingerprint recognition, and web crawling to penetrate the website. The vulnerability scanning process of the website using log records and HTML output helps users discover the vulnerability information of the website in a short time, patch the website in time. It can reduce the security risks caused by website vulnerabilities.
{"title":"Web Vulnerability Detection Analyzer Based on Python","authors":"Dawei Xu, Tianxin Chen, Zhonghua Tan, Fudong Wu, Jiaqi Gao, Yunfan Yang","doi":"10.4018/ijdcf.302875","DOIUrl":"https://doi.org/10.4018/ijdcf.302875","url":null,"abstract":"In the information age, hackers will use Web vulnerabilities to infiltrate websites, resulting in many security incidents. To solve this problem, security-conscious enterprises or individuals will conduct penetration tests on websites to test and analyze the security of websites, but penetration tests often take a lot of time. Therefore, based on the traditional Web vulnerability scanner, the Web vulnerability detection analyzer designed in this article uses vulnerability detection technologies such as sub-domain scanning, application fingerprint recognition, and web crawling to penetrate the website. The vulnerability scanning process of the website using log records and HTML output helps users discover the vulnerability information of the website in a short time, patch the website in time. It can reduce the security risks caused by website vulnerabilities.","PeriodicalId":44650,"journal":{"name":"International Journal of Digital Crime and Forensics","volume":"38 1","pages":"1-17"},"PeriodicalIF":0.7,"publicationDate":"2022-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88889838","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}
This paper proposes a monocular depth label propagation model, which describes monocular images into depth label distribution for the target classification matching. 1) Depth label propagation by hybrid sampling and salient region sifting, improve the discrimination of detection feature categories. 2) Depth label mapping and spectrum clustering to classify target, define the depth of the sorting rules. The experimental results of motion recognition and 3D point cloud processing, show that this method can approximately reach the performance of all previous monocular depth estimation methods. The neural network model black box training learning module is not used, which improves the interpretability of the proposed model.
{"title":"Monocular Depth Matching With Hybrid Sampling and Depth Label Propagation","authors":"Ye Hua, Qu Xi Long, Lihua Jin","doi":"10.4018/ijdcf.302879","DOIUrl":"https://doi.org/10.4018/ijdcf.302879","url":null,"abstract":"This paper proposes a monocular depth label propagation model, which describes monocular images into depth label distribution for the target classification matching. 1) Depth label propagation by hybrid sampling and salient region sifting, improve the discrimination of detection feature categories. 2) Depth label mapping and spectrum clustering to classify target, define the depth of the sorting rules. The experimental results of motion recognition and 3D point cloud processing, show that this method can approximately reach the performance of all previous monocular depth estimation methods. The neural network model black box training learning module is not used, which improves the interpretability of the proposed model.","PeriodicalId":44650,"journal":{"name":"International Journal of Digital Crime and Forensics","volume":"371 1","pages":"1-14"},"PeriodicalIF":0.7,"publicationDate":"2022-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76316161","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}
With the development of artificial intelligence, there are more and more applications related to face images. The recording of face information causes potential cyber security risks and personal privacy disclosure risks to the public. To solve this problem, we hope to protect face privacy through face anonymity. This paper designs a conditional autoencoder that uses the data preprocessing method of image inpainting. Based on the realistic generation ability of StyleGAN, our autoencoder model introduces facial pose information as conditional information. The input image only contains pre-processed face-removed images. Our method can generate high-resolution images and maintain the posture of the original face. It can be used for identity-independent computer vision tasks. Experiments further proves the effectiveness of our anonymization framework.
{"title":"Face Anonymity Based on Facial Pose Consistency","authors":"Junchang Wang","doi":"10.4018/ijdcf.302872","DOIUrl":"https://doi.org/10.4018/ijdcf.302872","url":null,"abstract":"With the development of artificial intelligence, there are more and more applications related to face images. The recording of face information causes potential cyber security risks and personal privacy disclosure risks to the public. To solve this problem, we hope to protect face privacy through face anonymity. This paper designs a conditional autoencoder that uses the data preprocessing method of image inpainting. Based on the realistic generation ability of StyleGAN, our autoencoder model introduces facial pose information as conditional information. The input image only contains pre-processed face-removed images. Our method can generate high-resolution images and maintain the posture of the original face. It can be used for identity-independent computer vision tasks. Experiments further proves the effectiveness of our anonymization framework.","PeriodicalId":44650,"journal":{"name":"International Journal of Digital Crime and Forensics","volume":"45 1","pages":"1-12"},"PeriodicalIF":0.7,"publicationDate":"2022-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82573149","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：