Community detection is essentially a clustering problem. The introduction of k-means algorithm into community detection research has many advantages in terms of practicability and algorithm performance. However, k-mea...
详细信息
ISBN:
(纸本)9781665420723
Community detection is essentially a clustering problem. The introduction of k-means algorithm into community detection research has many advantages in terms of practicability and algorithm performance. However, k-means algorithm has a fatal defect – it needs to determine the community division number K in advance, which is very difficult for community discovery without prior knowledge. In this paper, a three-stage community partitioning algorithm CDGE is proposed. In the first phase, nodes in the network are embedded into low-dimensional vectors using deepWalk embedding technology. In the second stage, plot the distance distribution histogram of nodes to obtain K through the Canopy algorithm; in the third stage, input K into k-means algorithm to obtain the specific community partition results. Experimental verification in real and artificial networks shows that the clustering effect and clustering speed of this algorithm are improved.
Driver’s drowsiness is one of the leading causes of traffic accidents. Drowsiness can be detected using brain waves (EEG) because it reflects the cognitive state of the brain. However, using EEG for drowsiness detect...
详细信息
ISBN:
(纸本)9781665442084
Driver’s drowsiness is one of the leading causes of traffic accidents. Drowsiness can be detected using brain waves (EEG) because it reflects the cognitive state of the brain. However, using EEG for drowsiness detection in vehicles is still impractical. The non-contact brain signal sensor NBM (Neuro-Bio Monitor), introduced by Freer Logic, has a potential to detect the driver’s drowsiness while driving due to its unobtrusive nature of brain signal sensing. In this study, a novel driver drowsiness detection algorithm was introduced using NBM signals. In addition, the feasibility of the NBM-based drowsiness detection algorithm in a driving simulator environment was evaluated by comparison with conventional methods including the eyelid closure ratio (PERCLOS) and the self-reported Karolinska Sleepiness Scale (KSS). The study recruited 14 healthy adults and asked them to drive about 70 minutes on a night drive mode. NBM signals were recorded simultaneously with PERCLOS and KSS. The algorithm shows an accuracy of 78.79% and a detection rate of 95% by comparison with KSS based drowsiness. It also shows a correlation of more than 70% with PERCLOS. This study demonstrated that NBM can be used in vehicle to detect driver drowsiness during driving.
At present, the traditional slope monitoring technology is relatively backward. Shadow detection has always been the difficulty and focus of this technology in slope monitoring. In recent years, digital image processi...
详细信息
ISBN:
(纸本)9781665415385;9781665447034
At present, the traditional slope monitoring technology is relatively backward. Shadow detection has always been the difficulty and focus of this technology in slope monitoring. In recent years, digital image processing technology has been developed rapidly, which provides the possibility to solve this difficult problem. Therefore, this paper proposes the shadow detection research in slope monitoring based on digital image processing. In this paper, the slope monitoring and image digitization technology in our country are deeply studied. The results show that the digital image measurement method has obvious advantages over the traditional slope monitoring method, and has improved the monitoring accuracy, operation difficulty and staff safety. According to the actual needs of shadow detection in slope monitoring, combined with the characteristics of digital image measurement method, the corresponding optimization adjustment is made. A shadow detection algorithm based on saturation is proposed. Compared with the traditional algorithm, the algorithm simplifies the calculation steps and improves the accuracy of calculation. In addition, in order to ensure the detection effect, this paper also establishes the corresponding detection indicators, and evaluates the detection results from the accuracy rate, false alarm rate and missing detection rate. In order to further verify the actual effect of this method, the corresponding comparative experiments are carried out. The experimental results show that the slope shadow detection model based on digital image calculation has higher detection accuracy than the traditional DGM detection model, and is more suitable for the current land environment of slope monitoring in China.
As a life-threatening disease, stroke can lead to long-term problems affecting the patients’ daily living ability. A common problem facing post-stroke patients is foot drop. An emerging modality of interest for corre...
详细信息
As a life-threatening disease, stroke can lead to long-term problems affecting the patients’ daily living ability. A common problem facing post-stroke patients is foot drop. An emerging modality of interest for correcting the foot drop is to combine both actuated ankle-foot orthosis (AAFO) and functional electrical stimulation (FES). Such hybrid assistive system not only ensure effective assistance but also can avoid fast muscular fatigue due to excessive muscular stimulation. Due to the significant changes in the ankle joint’s kinematics and kinetics with gait cycles, optimization control strategies for hybrid AAFO and FES systems are highly demanded. However, it is challenging to develop accurate gait phase detection algorithms to guide the control of AAFO and FES while ensuring robustness with respect to the diversity and variability of patients’ gaits. In this paper, we present a novel swing sub-phase detection algorithm based on a moving average convergence divergence (MACD) indicator. The proposed detection algorithm uses only information collected from the affected leg by means of two inertia measurement units (IMU) and the AAFO. Moreover, a gait-phase based control strategy is developed to optimize the assistive effect of a hybrid AAFO and FES system. Experimental results with five healthy show the potential of the proposed approaches in ensuring both satisfactory ankle joint trajectory tracking and effective reduction in stimulation intensity, compared to the use of conventional FES assistance.
We propose a GAN based soft failure detection and identification algorithm. For the training, the detection and identification only need normal samples and very few soft failure samples, respectively. 95% accuracy is ...
详细信息
ISBN:
(纸本)9781943580866
We propose a GAN based soft failure detection and identification algorithm. For the training, the detection and identification only need normal samples and very few soft failure samples, respectively. 95% accuracy is achieved in experiment.
There are some problems in the surface defect detection of industrial aluminum products, such as small defect samples, extreme length-to-width ratio of defect, low precision of small defect detection, etc. To solve th...
详细信息
ISBN:
(纸本)9781665411578
There are some problems in the surface defect detection of industrial aluminum products, such as small defect samples, extreme length-to-width ratio of defect, low precision of small defect detection, etc. To solve these problems, an aluminum surface defect detection algorithm is proposed based on improved Faster RCNN. The number of defect samples is increased by data augmentation, and the residual network ResNet50 is employed as the backbone feature extraction network to extract aluminum defect features. Then the path enhancement feature pyramid network (PAFPN) is added to the backbone feature extraction network to form a multi-scale feature map which strengthens the utilization of feature information from the lower layers. Soft non-maximum suppression (Soft-NMS) is used to further improve the detection performance of the algorithm. Results show that the mean average accuracy (mAP) of the proposed algorithm is 78.8%, which is 2.2% higher than the original algorithm.
The traditional collision detection of triangular patches is mostly attributed to the position relationship between line segments and points in triangular patches. To judge the position relationship between the three ...
详细信息
ISBN:
(纸本)9781665428781
The traditional collision detection of triangular patches is mostly attributed to the position relationship between line segments and points in triangular patches. To judge the position relationship between the three edges and three vertices of triangular patches, it is necessary to calculate the spatial position relationship for many times. The separation, intersection and parallel relationship between spatial triangular patches should be discussed in many cases, which is very complex. While the intersection judgment of triangles on the two-dimensional plane is much simpler. In view of this, this paper proposes a new collision detection method that transforms two spatial triangular patches into the same plane triangular patch. The complexity of triangle collision detection is effectively reduced and the efficiency of collision detection is improved.
Traditional pedestrian detection methods can no longer be effective improvement of pedestrian detection, so this paper proposes a pedestrian detection method combined with the attention mechanism, SE-Net attention mec...
详细信息
ISBN:
(数字)9781728180281
ISBN:
(纸本)9781728180298
Traditional pedestrian detection methods can no longer be effective improvement of pedestrian detection, so this paper proposes a pedestrian detection method combined with the attention mechanism, SE-Net attention mechanism and ECA-Net attention mechanism for channel focusing on sex to suppress some not important information channel, which greatly improves the recognition accuracy. Algorithm in this paper, based on CenterNet experiment, by predicting the pedestrian center and return the size information, this kind of anchor-free algorithm greatly reduces the time complexity. By targeting some occlusion and small scale pedestrian attention, our results on the ETH dataset are a good improvement over the original network.
The detection of of Heavy Hitter (HH) flows in a network device is a critical building block in many control and management tasks. A flow is considered a Heavy Hitter flow if its portion from the total traffic surpass...
详细信息
The detection of of Heavy Hitter (HH) flows in a network device is a critical building block in many control and management tasks. A flow is considered a Heavy Hitter flow if its portion from the total traffic surpasses a given threshold. One of the most important aspect of this detection is its practicality; i.e., being able to work in line rate using the available scarce local memory in the device. In this paper, we present a practical heavy hitters detection algorithm that requires a constant amount of memory (not related to the number of flows or the number of packets) and performs at most O(1) operation per packet to keep with line rate. We present an analysis of errors for our algorithm and compare it to state-of-the-art monitoring solutions, showing a superior performance where the allocated memory is less than 1 MB. In particular, we are able to detect more HH flows with less false positive without increasing the per-packet processing time.
Logic target attack and sequence target attack are two types of sophisticated and vicious cyber-attacks, which have not been paid a necessary attention in most critical part, i.e., the process layer network in digital...
详细信息
ISBN:
(纸本)9781665414401
Logic target attack and sequence target attack are two types of sophisticated and vicious cyber-attacks, which have not been paid a necessary attention in most critical part, i.e., the process layer network in digital substation. The relationships of such two types attack are explored, and networked structure of an intrusion detection & prevention system for the process layer network is proposed. A whitelist-based detection scheme is also proposed; firstly, the corresponding packets are described from the packet transfer path and the circuit breaker action packet; then, the different detection methods are used to check the packets; the simulation test results and verification show that the proposed method can detect an abnormal transmission path message and a replay attack with inserted circuit breaker action packet.
暂无评论