With the advent of cost-efficient depth cameras, many effective feature descriptors have been proposed for action recognition from depth sequences. However, most of them are based on single feature and thus unable to extract the action information comprehensively, e.g., some kinds of feature descriptors can represent the area where the motion occurs while they lack the ability of describing the order in which the action is performed. In this paper, a new feature representation scheme combining different feature descriptors is proposed to capture various aspects of action cues simultaneously. First of all, a depth sequence is divided into a series of sub-sequences using motion energy based spatial-temporal pyramid. For each sub-sequence, on the one hand, the depth motion maps (DMMs) based completed local binary pattern (CLBP) descriptors are calculated through a patch-based strategy. On the other hand, each sub-sequence is partitioned into spatial grids and the polynormals descriptors are obtained for each of the grid sequences. Then, the sparse representation vectors of the DMMs based CLBP and the polynormals are calculated separately. After pooling, the ultimate representation vector of the sample is generated as the input of the classifier. Finally, two different fusion strategies are applied to conduct fusion. Through extensive experiments on two benchmark datasets, the performance of the proposed method is proved better than that of each single feature based recognition method.
One of the most significant study directions is node positioning in wireless sensor networks (WSNs). Because the existing RSSI-based triangle centroid localization technique is susceptible to the surrounding environment, this paper proposes an improved triangle centroid localization algorithm based on point-in-triangulation (PIT) criterion (ITCL-PIT) in terms of positioning accuracy and response speed. When combined with the actual placement situation in conventional triangle centroid localization, the suggested algorithm considers the estimated coordinates of the junction points as extra beacon nodes. As a result of the new beacon nodes, the size of the triangle in the junction region is decreased. Then, using the PIT criteria, keep calculating until the predicted position of the node is outside the triangle. Finally, the unknown node's coordinates are determined using the centroid approach. Based on the guaranteed response time, when the communication distance is 15–30 m, the ITCL-PIT method may enhance localization accuracy by up to five times when compared to the standard triangle centroid localization approach. Furthermore, the proposed technique has higher localization accuracy and faster response speed than the centroid iterative estimation approach, and the response time of the ITCL-PIT algorithm is reduced by around 17%. In addition, the experimental platform is built to ensure that the proposed strategy effectively lowers positioning error.
From the perspective of ensuring life safety, combined with the advantages of high-speed time response and energy conservation of white light emitting diodes (LEDs), the visible light indoor positioning algorithm based on fire safety is proposed in the paper. First, the model is designed which needs three LED lights arranged in a straight line and positioned in the geographically north direction on the top of the model. Then, the proposed algorithm is discussed and analyzed when the camera is located at the center of the model and facing north, when the camera is located at the center of the model and the angle is rotated, and when the camera is located at any position of the model, respectively. It can accurately calculate the current position of the camera, its response speed is fast and the positioning accuracy is high. Furthermore, this paper also verifies the practicability and reliability of the algorithm by designing the visible light indoor positioning system based on fire safety rescue in natural environment and smoke environment. The experimental results show that the positioning error does not exceed 0.70 cm in smoke environment.
Considering that the traditional triangle centroid localization algorithm based on RSSI is susceptible to surrounding environment, this paper improves the algorithm from two aspects of positioning accuracy and response speed also proposes an improved triangle centroid localization algorithm based on PIT criterion. Combined with actual positioning situation, the algorithm treats the calculated coordinates of the intersection points as the new beacon nodes. Thus, the area of triangle in the intersection region is reduced. Repeat positioning process until the predicted position of node is outside the triangle according to the PIT criterion. Compared with traditional triangle centroid localization algorithm, it showed from the simulation results that the improved triangle centroid localization algorithm can increase the localization accuracy up to 5 times based on the guaranteed response time when communication distance is 15 ~ 30 m, and this algorithm has higher localization accuracy and faster response speed than centroid iterative estimation algorithm in larger communication range. In additions, the experimental platform is built to verify that the proposed algorithm can effectively reduce the positioning error.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.