摘要 :
In this work, a model has been developed for detecting popular tags belonging to suspicious group using shares of active hackers and followers on Twitter social network. Term frequency-inverse document frequency (tf-idf) is reinte...
展开
In this work, a model has been developed for detecting popular tags belonging to suspicious group using shares of active hackers and followers on Twitter social network. Term frequency-inverse document frequency (tf-idf) is reinterpreted with the number of favorite and re-tweet to detect popular tags belonging to suspicious group. The obtained feature space is used for detecting the most strongly suspected which are similar to the target hackers. The results show that suspected profiles, which are detected by our model, have been closed by Twitter with course decision.
收起
摘要 :
In this work, a model has been developed for detecting popular tags belonging to suspicious group using shares of active hackers and followers on Twitter social network. Term frequency-inverse document frequency (tf-idf) is reinte...
展开
In this work, a model has been developed for detecting popular tags belonging to suspicious group using shares of active hackers and followers on Twitter social network. Term frequency-inverse document frequency (tf-idf) is reinterpreted with the number of favorite and re-tweet to detect popular tags belonging to suspicious group. The obtained feature space is used for detecting the most strongly suspected which are similar to the target hackers. The results show that suspected profiles, which are detected by our model, have been closed by Twitter with course decision.
收起
摘要 :
Security awareness is becoming more and more critical in modern society. In this paper, we propose an infrared pedestrian tracking and trajectory analysis system. The system is able to monitor large area like factories and communi...
展开
Security awareness is becoming more and more critical in modern society. In this paper, we propose an infrared pedestrian tracking and trajectory analysis system. The system is able to monitor large area like factories and communities, and provides monito
收起
摘要 :
Security awareness is becoming more and more critical in modern society. In this paper, we propose an infrared pedestrian tracking and trajectory analysis system. The system is able to monitor large area like factories and communi...
展开
Security awareness is becoming more and more critical in modern society. In this paper, we propose an infrared pedestrian tracking and trajectory analysis system. The system is able to monitor large area like factories and communities, and provides monito
收起
摘要 :
Security awareness is becoming more and more critical in modern society. In this paper, we propose an infrared pedestrian tracking and trajectory analysis system. The system is able to monitor large area like factories and communi...
展开
Security awareness is becoming more and more critical in modern society. In this paper, we propose an infrared pedestrian tracking and trajectory analysis system. The system is able to monitor large area like factories and communities, and provides monitoring and suspicious alarm functions. An attention based object detector is structured to integrate features from different level. Deep-sort style tracker is used by following the principle of tracking-by-detection. Based on the tracking results, the system provides the function of suspect alarm by analyzing the trajectories. The performance of pedestrian tracking and suspect alarm satisfies the industrial needs in the experiment.
收起
摘要 :
The suspect vehicle detection system normally compares the list of criminal license plates and vehicle license plates gathering from various sensors in order to identify the criminal vehicles or the suspect vehicles. However, the ...
展开
The suspect vehicle detection system normally compares the list of criminal license plates and vehicle license plates gathering from various sensors in order to identify the criminal vehicles or the suspect vehicles. However, the traditional process of comparing those license plates utilizing the matching of alphabet character is not effective. If the characters do not match any one character, the system can not detect the criminal vehicles or the suspect vehicles. This paper proposes the use of reputation algorithm to detect the criminal vehicles crossing the checkpoint whose license plates match the blacklist. In addition to that, we use association analysis concept to detect the suspect vehicles that have ever passed the checkpoint that may be related to the criminal activity records. Our method can detect the suspect vehicles with fake license plate by using color, brand and type of the vehicles instead of only the license plate matching to the blacklists. These two techniques use a blacklist of criminal vehicles and criminal activity recorded in a criminal report database of Defence Technology Institute (DTI), Thailand, to help facilitate the detection process. The result shows that the reputation algorithm and the association analysis concept can improve the detection capability of the suspect vehicle detection system.
收起
摘要 :
Automatic polyp detection during colonoscopy is beneficial for reducing the risk of colorectal cancer. However, due to the various shapes and sizes of polyps and the complex structures in the intestinal cavity, some normal tissues...
展开
Automatic polyp detection during colonoscopy is beneficial for reducing the risk of colorectal cancer. However, due to the various shapes and sizes of polyps and the complex structures in the intestinal cavity, some normal tissues may display features similar to actual polyps. As a result, traditional object detection models are easily confused by such suspected target regions and lead to false-positive detection. In this work, we propose a multi-branch spatial attention mechanism based on the one-stage object detection framework, YOLOv4. Our model is further jointly optimized with a top likelihood and similarity to reduce false positives caused by suspected target regions. A similarity loss is further added to identify the suspected targets from real ones. We then introduce a Cross Stage Partial Connection mechanism to reduce the parameters. Our model is evaluated on the private colonic polyp dataset and the public MICCAI 2015 grand challenge dataset including the CVC-Clinic 2015 and Etis-Larib, both of the results show our model improves performance by a large margin and with less computational cost.
收起
摘要 :
Automatic polyp detection during colonoscopy is beneficial for reducing the risk of colorectal cancer. However, due to the various shapes and sizes of polyps and the complex structures in the intestinal cavity, some normal tissues...
展开
Automatic polyp detection during colonoscopy is beneficial for reducing the risk of colorectal cancer. However, due to the various shapes and sizes of polyps and the complex structures in the intestinal cavity, some normal tissues may display features similar to actual polyps. As a result, traditional object detection models are easily confused by such suspected target regions and lead to false-positive detection. In this work, we propose a multi-branch spatial attention mechanism based on the one-stage object detection framework, YOLOv4. Our model is further jointly optimized with a top likelihood and similarity to reduce false positives caused by suspected target regions. A similarity loss is further added to identify the suspected targets from real ones. We then introduce a Cross Stage Partial Connection mechanism to reduce the parameters. Our model is evaluated on the private colonic polyp dataset and the public MICCAI 2015 grand challenge dataset including the CVC-Clinic 2015 and Etis-Larib, both of the results show our model improves performance by a large margin and with less computational cost.
收起
摘要 :
Although rapid malware detection is very important, the detection is difficult due to the increase of new malware. In recent years, blockchain technology has attracted the attention of many people due to its four main characterist...
展开
Although rapid malware detection is very important, the detection is difficult due to the increase of new malware. In recent years, blockchain technology has attracted the attention of many people due to its four main characteristics of decentralization, persistency, anonymity, and auditability. In this paper, we propose a blockchain-based malware detection method that uses shared signatures of suspected malware files. The proposed method can share the signatures of suspected files between users, allowing them to rapidly respond to increasing malware threats. Further, it can improve the malware detection by utilizing signatures on the blockchain. In the evaluation experiment, we perform a more real simulation compared with our previous work to evaluate the detection accuracy. Compared with heuristic methods or behavior-based methods only, the proposed system which uses these methods plus signature-based method using shared signatures on the blockchain improved the false negative rate and the false positive rate.
收起
摘要 :
The crime problems become critical issues for national security especially the security of border and intelligent transportation systems (ITSs). These affect the economy, investment, tourism, and society. As a result, the automati...
展开
The crime problems become critical issues for national security especially the security of border and intelligent transportation systems (ITSs). These affect the economy, investment, tourism, and society. As a result, the automatic suspect vehicle detection emerges as one of effective tools to tackle the problems. However, the traditional process normally uses criminal vehicle data in blacklist comparing with vehicle data gathering from various sensors. This comparison is not effective and accurate that might be from not up-to-date data in the blacklist. Sometimes the blacklist is not available. This paper proposes the criminal behavior analysis method to detect suspect vehicles that are potentially involved in criminal activity. It must not rely on the blacklist. The analysis is conditional on journey path and the involvement of criminal activities. In additional, public officials believe that the suspect vehicle will choose the journey path without a checkpoint. Therefore, we used the journey path analysis techniques together with the association rule mining to analyze such criminal behavior. From extensive experiments, the results show that the proposed method can increase the suspect detection accuracy rate 17.24% beyond the traditional counterpart.
收起
原文获取