[1]高 强,潘 俊,洪锐锋.基于 CNN 的机场安检危险品自动识别研究[J].计算机技术与发展,2019,29(10):95-99.[doi:10. 3969 / j. issn. 1673-629X. 2019. 10. 020]
　GAO Qiang,PAN Jun,HONG Rui-feng.Research on Automatic Recognition of Dangerous Goods in Airport Security Inspection Based on CNN[J].,2019,29(10):95-99.[doi:10. 3969 / j. issn. 1673-629X. 2019. 10. 020]
点击复制

基于 CNN 的机场安检危险品自动识别研究()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
29
期数:
2019年10期
页码:
95-99
栏目:
应用开发研究
出版日期:
2019-10-10

文章信息/Info

Title:
Research on Automatic Recognition of Dangerous Goods in Airport Security Inspection Based on CNN
文章编号:
1673-629X(2019)10-0095-0
作者:
高 强; 潘 俊; 洪锐锋
广州民航职业技术学院 计算机系,广东 广州 510403
Author(s):
GAO Qiang; PAN Jun; HONG Rui-feng
Department of Computer,Guangzhou Civil Aviation College,Guangzhou 510403,China
关键词:
危险品; CNN; 自动识别; 不均衡; GAN
Keywords:
dangerous goods; CNN; automatic recognition; imbalance; GAN
分类号:
TP391.41
DOI:
10. 3969 / j. issn. 1673-629X. 2019. 10. 020
摘要:
机场安检是民航安全飞行的重要保障。 针对机场安检危险品人工识别工作量大、效率低、易疲劳误判及危险品图像数据集不均衡导致识别准确率低的问题,提出一种基于 GAN 数据增强的卷积神经网络危险品自动识别模型。 首先利用GAN 使危险品图像数据集均衡化,然后将图像输入由 4 个卷积层、1 个全连接层构成的卷积神经网络模型进行训练,训练时引入随机失活优化技术,使模型得到更好的识别效果。 在 2017 公安一所危险品图像数据集上的实验结果表明,经过均衡化处理后,模型的识别准确率达到 90.7%,较均衡化之前提高了 33.4%。 另外,经过对比实验,模型的识别准确率分别比 GoogLeNet、AlexNet、ResNet 高出 5.8%、7.2%和 5.4%。 该模型具有较高的识别准确率及较好的实时性,对提升机场安检智能化水平具有积极意义。
Abstract:
Airport security inspection is an important guarantee for civil aviation safety flight. According to the problems of heavy workload,low efficiency,easy fatigue misjudgment with artificial recognition and low recognition accuracy caused by imbalance of dangerous goods image dataset in airport security inspection,we propose a convolution neural network automatic recognition model for dangerous goods based on oversampling. Firstly,the GAN is used to equalize the dataset of dangerous goods image,and then the image is inputted into the convolution neural network model composed of four convolution layers and one full-connection layer for training. The stochastic deactivation optimization technique is introduced in the training to get better recognition effect. The experimental results on a dangerous goods image dataset of public security in 2017 show that the recognition accuracy of the model can reach 90.7% after equalization,which is 33.4% higher than that before equalization. In addition,the recognition accuracy of the model is 5.8%, 7.2% and 5.4% higher than that of GoogLeNet,AlexNet and ResNet respectively. The model has high recognition accuracy and great real-time performance,which is of positive significance to improve the level of airport security intelligence.

相似文献/References:

[1]李圣普,王小辉. 安全优先运输路径选择算法研究与仿真[J].计算机技术与发展,2015,25(09):94.
　LI Sheng-pu,WANG Xiao-hui. Simulation and Research of Explosive Goods Transportation Route Choice Algorithm[J].,2015,25(10):94.
[2]赵胜,赵学健,张欣慧,等.基于Sobel 算子和 CNN 的车辆轨迹识别系统[J].计算机技术与发展,2018,28(07):169.[doi:10.3969/ j. issn.1673-629X.2018.07.036]
　ZHAO Sheng,ZHAO Xue-jian,ZHANG Xin-hui,et al.Vehicle Trajectory Recognition System Based on Sobel Operator and CNN[J].,2018,28(10):169.[doi:10.3969/ j. issn.1673-629X.2018.07.036]
[3]任飞凯,邱晓晖.基于 LBP 和数据扩充的 CNN 人脸识别研究[J].计算机技术与发展,2020,30(03):62.[doi:10. 3969 / j. issn. 1673-629X. 2020. 03. 012]
　REN Fei-kai,QIU Xiao-hui.Research on Face Recognition of CNN Based on LBP and Data Expansion[J].,2020,30(10):62.[doi:10. 3969 / j. issn. 1673-629X. 2020. 03. 012]
[4]王新美,丁爱玲,雷梦宁,等.基于 CNN 和 SVM 融合的交通标志识别[J].计算机技术与发展,2020,30(06):7.[doi:10. 3969 / j. issn. 1673-629X. 2020. 06. 002]
　WANG Xin-mei,DING Ai-ling,LEI Meng-ning,et al.Traffic Sign Recognition Based on Combination of CNN and SVM[J].,2020,30(10):7.[doi:10. 3969 / j. issn. 1673-629X. 2020. 06. 002]
[5]何烨辛,谷　林,孙　晨.基于ＣＮＮ的程序编译错误信息特征提取[J].计算机技术与发展,2021,31(05):204.[doi:10. 3969 / j. issn. 1673-629X. 2021. 05. 035]
　,ＣＮＮ－ｂａｓｅｄＰｒｏｇｒａｍ ＣｏｍｐｉｌａｔｉｏｎＥｒｒｏｒＭｅｓｓａｇｅＦｅａｔｕｒｅＥｘｔｒａｃｔｉｏ[J].,2021,31(10):204.[doi:10. 3969 / j. issn. 1673-629X. 2021. 05. 035]
[6]苏魁麟,张 凯,吕学强,等.基于融合模型的名词隐喻识别[J].计算机技术与发展,2022,32(06):192.[doi:10. 3969 / j. issn. 1673-629X. 2022. 06. 032]
　SU Kui-lin,ZHANG Kai,LYU Xue-qiang,et al.Noun Metaphor Recognition Based on Fusion Model[J].,2022,32(10):192.[doi:10. 3969 / j. issn. 1673-629X. 2022. 06. 032]

更新日期/Last Update: 2019-10-10