Xuhua Pan

Xuhua Pan

Auto-generated profile, based on publicly available information

Xuhua Pan

School of Information Engineering, Tianjin University of Commerce, Tianjin 300134, China

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
31 July 2021 in Scientific Programming
Verified
Journal Article
Open access
PDF file
LZLiyi Zhang
Liyi Zhang
MFMingyue Fu
Mingyue Fu
TFTeng Fei
Teng Fei
XPXuhua Pan
Xuhua Pan
Green development is the hot spot of cold chain logistics today. Therefore, this paper converts carbon emission into carbon emission cost, comprehensively considers cargo damage, refrigeration, carbon emission, time window, and other factors, and establishes the optimization model of location of low-carbon cold chain logistics in the Beijing-Tianjin-Hebei metropolitan area. Aiming at the problems of the fish swarm algorithm, this paper makes full use of the fireworks algorithm and proposes an improved fish swarm algorithm on the basis of the fireworks algorithm. By introducing the explosion, Gaussian mutation, mapping and selection operations of the fireworks algorithm, the local search ability and diversity of artificial fish are enhanced. Finally, the modified algorithm is applied to optimize the model, and the results show that the location scheme of low-carbon cold chain logistics in Beijing-Tianjin-Hebei metropolitan area with the lowest total cost can be obtained by using fireworks-artificial fish swarm algorithm.

13 September 2022 in Scientific Programming
Verified
Journal Article
Open access
BYBai Yu
Bai Yu
XPXuhua Pan
Xuhua Pan
XLXuefeng Li
Xuefeng Li
GLGaohua Liu
Gaohua Liu
YMYunpeng Ma
Yunpeng Ma
Feature pyramid network is widely used in advanced object detection. By simply changing the network connection, the performance of small object detection can be greatly improved without increasing the amount of calculation of the original model. However, the algorithm still has some shortcomings. Therefore, a new attention feature pyramid network (attention PFN) is proposed. Firstly, an improved receptive field module is added, which can make full use of global and local context information. Secondly, the connection mode of the pyramid is further optimized. Deconvolution is used to replace the nearest neighbor interpolation in top-down up-sampling and a channel attention module is added to the horizontal connection to highlight important context information. Finally, adaptive fusion and spatial feature balance are used for each feature pyramid, so that the network can learn the weights of different feature layers. Each pyramid layer contains more discrimination information. Attention PFN is tested on Pascal VOC and MS COCO datasets, respectively. The experiment results revealed that the proposed method has better performance than the original algorithm. Therefore, the attention PFN is an effective algorithm.