In many practical applications like autonomous driving and robot navigation, the large-scale 3D point cloud segmentation method is required to be both fast and efficient. In this paper, a fast and efficient large-scal...
详细信息
In many practical applications like autonomous driving and robot navigation, the large-scale 3D point cloud segmentation method is required to be both fast and efficient. In this paper, a fast and efficient large-scale point cloud semantic segmentation network, namely FFA-Net, is proposed. We used U-Net architecture as the backbone architecture, and used the random sampling algorithm to down sample the 3D points faster. To maintain the speed advantage of randomsampling and protect integral information of 3D point cloud, we designed a lightweight operator, called fast feature aggregation (FFA) operator, to learn local features of 3D point cloud efficiently, and we equipped this FFA operator in a dilated residual block to aggregate local features within a larger receptive field hierarchically. This operator contains only minimal of learnable parameters, which makes the segmentation network not only improved in segmentation performance, but also in computation speed. Extensive experimental results on three large-scale 3D point cloud benchmarks have verified the effectiveness of our method in both segmentation performances and speed.
Vehicle-to-grid (V2G) is an important component of smart grids and plays a significant role in improving grid stability, reducing energy consumption and generating cost. However, while electric vehicles are being char...
详细信息
Vehicle-to-grid (V2G) is an important component of smart grids and plays a significant role in improving grid stability, reducing energy consumption and generating cost. However, while electric vehicles are being charged, it is possible to expose the location and movement trajectories of the electric vehicles, thereby triggering a series of privacy and security issues. In response to this problem, we propose a new quadtree-based spatial decomposition algorithm to protect the location privacy of electric vehicles. First of all, we use a random sampling algorithm, which is based on differential privacy, to obtain enough spatial data to achieve the balance between large-scale spatial data and the amount of noise. Secondly, in order to overcome the shortcomings of using tree height to control Laplacian noise in the quadtree, we use sparse vector technology to control the noise added to the tree nodes. Finally, according to the vehicle-to-grid network structure in the smart grid, we propose a location privacy protection model based on distributed differential privacy technology for EVs in vehicle-to-grid networks. We demonstrate application of the proposed model in real spatial data and show that it can achieve the best effect on the security of the algorithm and the availability of data.
This paper proposes a method for inferring the road network from Global Position System (GPS) traces, which is composed of intersections and the roads between each pair of directly-connected intersections. random Samp...
详细信息
ISBN:
(纸本)9781479960781
This paper proposes a method for inferring the road network from Global Position System (GPS) traces, which is composed of intersections and the roads between each pair of directly-connected intersections. randomsampling (RANSAC) algorithm is used to cluster the turning points, where the users change their moving directions, into intersections. All of the GPS traces are segmented by the intersections, resulting in connectivity matrix of the intersections and small GPS tracks for each pair of directly-connected intersections. At last, the road between each two directly-connected intersections is extracted through aligning and averaging all of the tracks using Dynamic Time Warping (DTW) algorithm. The main novelty of our methods is aligning the tracks point by point for each road using a "stretching and compression" strategy, which not only allows road estimation by averaging the aligned tracks, but also a deeper statistical analysis using their time alignment, such as analyzing the users' speed stability at a specific location. The experimental results show that our algorithm outperforms other methods by producing clean road network without spurious edges.
This paper introduces the problem of randomsampling from time-based sliding windows over weighted streaming data and presents a priority randomsampling (PRS) algorithm for this problem. The algorithm extends classic...
详细信息
ISBN:
(纸本)9781595934802
This paper introduces the problem of randomsampling from time-based sliding windows over weighted streaming data and presents a priority randomsampling (PRS) algorithm for this problem. The algorithm extends classic reservoir-samplingalgorithm and weighted random sampling algorithm with a reservoir to deal with the expiration of data items from time-based sliding window, and can avoid drawbacks of classic reservoir-samplingalgorithm and weighted samplingalgorithm with a reservoir. In the new algorithm, a key is assigned for each data item in the time-based sliding window by compromising its weight and arrival time, and works even when the number of data items in a sliding window varies dynamically over time. The experiments show that PRS algorithm is somewhat superior to WRS algorithm.
This paper discusses the random sampling algorithm for landmark windows over weighted streaming data, and presents a new algorithm by improving weighted randomsampling (WRS) algorithm with a reservoir. When a new dat...
详细信息
ISBN:
(纸本)9781424467129
This paper discusses the random sampling algorithm for landmark windows over weighted streaming data, and presents a new algorithm by improving weighted randomsampling (WRS) algorithm with a reservoir. When a new data item v(i) with weight w(i) arrives, a random number u(i) is generated, and a key k(i) is calculated by w(i) and u(i) for the data item. We maintain a candidate sample set by the keys of data items, and the keys of older data items is decreased periodically. The theoretic analysis and experiments show that the algorithm is effective and efficient for continuous data streams processing.
This paper introduces the problem of randomsampling from time-based sliding windows over data streams and presents a basic-window based priority-sample(BWPS) algorithm for this *** the BWPS algorithm,a key is calcula...
详细信息
ISBN:
(纸本)9780972147903
This paper introduces the problem of randomsampling from time-based sliding windows over data streams and presents a basic-window based priority-sample(BWPS) algorithm for this *** the BWPS algorithm,a key is calculated for each data item in the time-based sliding window,and data items with larger keys are selected to enter the sample to replace the data items with smaller *** algorithm extends the priority-sampling(PS) algorithm and achieves a significant improvement on efficiency,and has similar performance on storage comparing with PS *** can work well even when the number of data items in a sliding window varies dynamically over *** experiments show that the BWPS algorithm is effective and efficient for processing data items from sliding windows over data streams.
暂无评论