[1]许 哲,张少帅,郭 璐,等.无人机深度学习去雾算法[J].机械与电子,2021,(04):13-16.
 XU Zhe,ZHANG Shaoshuai,GUO Lu,et al. Deep Learning Defogging Algorithm for UAV[J].Machinery & Electronics,2021,(04):13-16.
点击复制

无人机深度学习去雾算法()
分享到:

机械与电子[ISSN:1001-2257/CN:52-1052/TH]

卷:
期数:
2021年04期
页码:
13-16
栏目:
设计与研究
出版日期:
2021-04-24

文章信息/Info

Title:
 Deep Learning Defogging Algorithm for UAV
文章编号:
1001-2257(2021)04-0013-04
作者:
 许 哲1张少帅2郭 璐34黄 鹤2王会峰2尹康迪2
1. 中国电子科技集团公司第20研究所,陕西 西安 710068;
2. 长安大学电子与控制工程学院,陕西 西安 710064;
3.西安爱生技术集团公司,陕西 西安 710075;
4. 西北工业大学无人机系统国家工程研究中心,陕西 西安 710072
Author(s):
 XU Zhe1ZHANG Shaoshuai2GUO Lu34HUANG He2WANG Huifeng2YIN Kangdi2
1. The 20th Research Institute, Xi’an 710068;

2.School of Electronics and Control Engineering, Chang’an University, Xi’an 710064;

3. Xi’an ASN Technology Group Company, Xi’an 710075,China;

4. UAV National Engineering Research Center,Northwestern Polytechnical University,Xi’an 710072,China

关键词:
 无人机去雾深度学习特征提取DLDN
Keywords:
unmanned aerial vehicle (UAV)defoggingdeep learningfeatureextractionDLDN
分类号:
TP391.41
文献标志码:
A
摘要:
 为解决无人机实际飞行过程中受到雾霾的影响,使获取的图像明显降质问题,提出了一种深度学习神经网络模型DLDN,利用神经网络模型实现无人机深度学习去雾算法。首先结合暗通道先验进行去雾网络模型设计,在保证去雾精度和效果下对网络结构进行优化;然后对含雾图像间隔多尺度卷积特征提取,采取最大池化层的方式对特征降维;最后通过修正线性单元和全连接对进行处理,得到回归结果。实验结果表明,该算法与原图相似度最高、鲁棒性较好且去雾效果明显,满足实际工程实时性要求。
Abstract:
 In order to solve the problem that the uav is affected by haze during actual flight and the acquired image is obviously degraded, a deep learning neural network model DLDN is proposed, and the neural network model is used to realize the deep learning defogging algorithm of UAV.Firstly, the defogging network model is designed based on the dark channel prior, and the network structure is optimized to ensure the accuracy and effect of defogging.Then multi-scale convolution feature is extracted from fog-containing image interval and the feature dimension is reduced by maximum pooling layer.Finally, the regression results are obtained by modifying the linear element and the full join pair.The experimental results show that the algorithm has the highest similarity with the original image, good robustness and obvious defogging effect, and meets the real-time requirements of practical engineering.

参考文献/References:

[1] SINGH D, KUMAR V. Single image haze removal using integrated dark and bright channel prior [J]. Modern physics letters B, 2018: 1850051.

[2] FATTAL R. Single image dehazing [J]. ACM Transactions on graphics, 2008, 27(3): 721-729.

[3] 黄鹤, 李昕芮, 宋京,等. 多尺度窗口的自适应透射率修复交通图像去雾方法[J]. 中国光学, 2019, 12(6): 1311-1320.

[4] 黄鹤,宋京,杜晶晶,等. 一种含雾交通图像梯度双边滤波算法[J]. 哈尔滨工程大学学报, 2018,39(10):1709-1714.

[5] NARASIMHAN S G, NAYAR S K. Contrast restoration of weather degraded images [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003, 25(6): 713-724.

[6] 黄鹤,宋京,王会峰,等.雾霾天气下基于二次滤波的交通图像去雾算法[J]. 科学技术与工程, 2016, 30(16): 274-277.

[7] 黄鹤, 宋京, 郭璐, 等. 基于新的中值引导滤波的交通视频去雾算法[J]. 西北工业大学学报,2018,36(3):414-419.

[8] HUANG H, SONG J,GUO L,et al. Haze removal method based on a variation function and colour attenuation prior for UAV remote-sensing images [J]. Journal of modern optics, 2019, 66(12): 1282-1295.

相似文献/References:

[1]雷永涛,周 权,张静静. 基于印制电路连接器变形问题的分析和设计改进[J].机械与电子,2019,(01):7.
 ,analysis and design improvement of deformation problem based on printed circuit connector[J].Machinery & Electronics,2019,(04):7.
[2]孟繁萃1,张 达2,张 明1. 自然风力对绝缘子结构的影响[J].机械与电子,2019,(01):3.
 ,the influence of natural wind on insulator structure[J].Machinery & Electronics,2019,(04):3.
[3]任智勇1,邓晓政1,马丁峰2. 一种发动机试飞数据快速处理方法[J].机械与电子,2019,(01):11.
 ,A Fast Processing Method of Multi-subject Engine Flight Test Data[J].Machinery & Electronics,2019,(04):11.
[4]冷雪锋,吴正明. 基于STC12C5A60S2的100W 小型风力发电系统设计[J].机械与电子,2019,(01):15.
 .The Design of 100W Small Wind Power Generation System Based on STC12C5A60S2[J].Machinery & Electronics,2019,(04):15.
[5]查金水,朱志远,邓友银,等. 某大型相控阵雷达测试系统结构设计[J].机械与电子,2019,(01):19.
 ,,et al.Structural Design of a Large Phased Array Radar Test System[J].Machinery & Electronics,2019,(04):19.
[6]万 欢,李伟达,李 娟. 一种面向上肢康复的便携式电刺激器设计[J].机械与电子,2019,(01):33.
 ,Design of Portable Electric Stimulator for Upper Limb Rehabilitation[J].Machinery & Electronics,2019,(04):33.
[7]郑锦汤1,李玉忠1,2. 紧凑型纯电动汽车动力系统匹配与性能仿真[J].机械与电子,2019,(01):38.
 ,Matching and Performance Simulation of Dynamic Systems of Compact Electric Vehicle[J].Machinery & Electronics,2019,(04):38.
[8]吕明珠1,2,苏晓明 1,等. 改进粒子群算法优化的支持向量机在滚动轴承故障诊断中的应用[J].机械与电子,2019,(01):42.
 ,,et al.Application of SVM Optimized by IPSO in Rolling Bearing Fault Diagnosis[J].Machinery & Electronics,2019,(04):42.
[9]韩雪莹1,张佳炜2,刘 梦1,等. 集中控制协调多个并网逆变器的谐波补偿[J].机械与电子,2019,(01):49.
 ,,et al.Centralized Control Coordination of Harmonic Compensation for Multiple Grid-connected Inverters[J].Machinery & Electronics,2019,(04):49.
[10]蒋 波1,3,周自强2,等. 基于PLC的报废汽车拆解线作业控制系统[J].机械与电子,2019,(01):54.
 ,,et al.PLC-based ELV Disassembly Line Operation Control System[J].Machinery & Electronics,2019,(04):54.

备注/Memo

备注/Memo:
收稿日期:2020-08-24

基金项目:装备预研领域基金(61403120105);陕西省自然科学基础研究计划面上项目(2019JM-610);陕西省创新人才推进计划-青年科技新星项目(2019KJXX-028);长安大学“省级大学生创新创业训练计划”项目资助(S202010710384)

作者简介:许 哲(1979—),男,河南洛阳人,博士,高级工程师,主要研究方向为无人系统设计,通信技术等;张少帅(1995-),男,陕西富平人,硕士研究生,主要研究方向为人工智能;郭 璐(1979—),女,陕西西安人,博士,高级工程师,主要研究方向为无人系统设计,信息融合等,通信作者;黄 鹤(1979-),男,副教授,研究方向为图像处理与嵌入式开发、无人机测控等。

更新日期/Last Update: 2021-04-15