Recurrent neural network multi-label aerial images classification

Ke-jun CHEN; Ye ZHANG

doi:10.3788/OPE.20202806.1404

您当前的位置：

首页 >

文章列表页 >

Recurrent neural network multi-label aerial images classification

更新时间：2020-07-13

- Recurrent neural network multi-label aerial images classification
- Optics and Precision Engineering Vol. 28, Issue 6, Pages: 1404-1413(2020)
- 作者机构：
  
  1.中国科学院大学长春光学精密机械与物理研究所应用光学国家重点实验室，吉林长春 130033
  2.中国科学院大学，北京 100039
- 作者简介：
- 基金信息：
- DOI：10.3788/OPE.20202806.1404
  CLC： TP391.7
- Received：02 December 2019，
  
  Accepted：21 January 2020，
  
  Published：15 June 2020
- 稿件说明：
移动端阅览
Ke-jun CHEN, Ye ZHANG. Recurrent neural network multi-label aerial images classification[J]. Optics and precision engineering, 2020, 28(6): 1404-1413.
DOI：

Ke-jun CHEN, Ye ZHANG. Recurrent neural network multi-label aerial images classification[J]. Optics and precision engineering, 2020, 28(6): 1404-1413. DOI： 10.3788/OPE.20202806.1404.

摘要

由于航空图像背景复杂，包含的物体类别多样，航空图像分类任务仍然面临困难。针对传统航空图像多标签分类算法准确率低、泛化性差的问题，本文提出了一种基于循环神经网络多标签航空图像分类方法。首先，采用超像素分割获取图像的低层特征，通过注意力机制生成注意力特征图；接着，采用交叉验证的方式获取最佳的图像尺度，将多尺度注意力特征图嵌入卷积神经网络中对图像进行特征提取；最后，采用改进的双向长短期记忆网络挖掘标签之间的相关性，改进的双向长短期记忆网络增加了输入门到输出门之间的连接，使输入状态可以更好地控制每一内存单元输出的信息，并且将遗忘门和输入门合并成单一的更新门，使得改进的双向长短期记忆网络可以学到更长时期的历史信息。结果显示，在图像变换尺度为1，1.3，2时，模型在UCM多标签数据集上的精确率和召回率分别达到了85.33%和87.05%，F1值达到了0.862。本文方法相比于原始VGGNet16模型，精确率提高了7.25%，召回率提高了8.94%。实验表明，该方法可以有效提高航空图像多标签分类任务的准确率。

Abstract

Due to the complexity of the background in aerial images and the diversity of object categories

aerial image classification is a challenging task. In order to address the problems of low accuracy and poor generalization in traditional multi-label aerial image classification methods

a method based on recurrent neural networks was proposed.In this method

the super-pixel segmentation algorithm was first used to obtain the low-level features of the image from which an attention map was generated. Subsequently

the best image scale was obtained by cross-validation

and multi-scale attention feature graphs were embedded into aconvolutional neural network in order to extract the features of the image.Finally

tomine the correlation between labels

an improved bidirectional Long Short-Term Memory (LSTM)network was proposed

which increases the connection from the input gate to the output gate

so that the input state can efficiently control the output information of each memory unit. The forget gate and the input gate were combined into a single update gate so that the improved bidirectional LSTM network can learn long-term historical information. The results obtained by applying the proposed method to the UCM multi-label dataset indicate that for scale values of 1

1.3

and 2

the accuracy and recall rates of the model are 85.33% and 87.05% respectively

while the F1 score reached 0.862. The accuracyand recall rates are found to be higher than those of theVGGNet16 model by 7.25% and 8.94% respectively.The experimental results thus indicate that the proposed method can effectively increase the accuracy of multi-label aerial image classification.

关键词

Keywords

references

郑远攀, 李广阳, 李晔.深度学习在图像识别中的应用研究综述[J].计算机工程与应用, 2019, 55(12):20-36.

ZHENG Y P, LI G Y, LI Y. Survey of application of deep learning in image recognition[J]. Computer Engineering and Applications , 2019, 55(12):20-36. (in Chinese)

李晓斌, 江碧涛, 王生进.光学遥感图像场景分类技术综述和比较[J].无线电工程, 2019, 49(4): 265-271.

LI X B, JIAN B T, WANG SH J. A review and comparison of optical remote sensing scene classification[J]. Radio Engineering , 2019, 49(4): 265-271. (in Chinese)

邓凌云.遥感图像分类中的遗传算法LVQ神经网络运用[J].现代电子技术, 2020, 43(1): 40-43.

DENG L Y. Application of LVQ neural network in remote sensing image classification[J]. Modern Electronics Technique , 2020, 43(1):40-43. (in Chinese)

杨州, 慕晓冬, 王舒洋, 等.基于多尺度特征融合的遥感图像场景分类[J].光学精密工程, 2018, 26(12): 232-240.

YANG ZH, MU X D, WANG SH Y, et al .. Scene classification of remote sensing images based on multi-scale features fusion[J]. Opt. Precision Eng ., 2018, 26(12): 232-240. (in Chinese)

李宇, 刘雪莹, 张洪群, 等.基于卷积神经网络的光学遥感图像检索[J].光学精密工程, 2018, 26(1): 200-207.

LI Y, LIU X Y, ZHANG H Q, et al .. Optical remote sensing image retrieval based on convolution neural network[J]. Opt. Precision Eng ., 2018, 26(1): 200-207. (in Chinese)

ZEGGADA A, BENBRAIKA S, MELGANI F, et al .. Multi-label conditional random field classification for UAV images[J]. IEEE Geoscience and Remote Sensing Letters , 2018, 15(3): 399-403.

KODA S, ZEGGADA A, MELGANI F, et al .. Spatial and structured SVM for multi-label image classification[J]. IEEE Transactions on Geoscience and Remote Sensing , 2018, 56(10): 5948-5960.

边小勇, 费雄君, 穆楠.基于尺度注意力网络的遥感图像场景分类[J].计算机应用, 2020, 40(3):872-877.

BIAN X Y, FEI X J, MU N. Remote sensing image scene classification based scale-attention network[J]. Journal of Computer Applications, 2020, 40(3):872-877. (in Chinese)

ACHANTA R, SHAJI A, SMITH K, et al .. SLIC superpixel compared to state-of-the-art superpixel methods[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence , 2012, 34(11): 2274-2282.

HE K, ZHANG X, REN S, et al .. Deep residual learning for image recognition[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2016: 770-778.

IOFFE S, SZEGEDY C. Batch normalization: Accelerating deep network training by reducing internal covariate shift[J]. Arxiv Preprint Arxiv: 1502.03167, 2015.

CHAUDHURI B, DEMIR B, CHAUDHURI S, et al .. Multi-label remote sensing image retrieval using a semi-supervised graph-theoretic method[J]. IEEE Transactions on Geoscience and Remote Sensing , 2018, 56(2):1144-1158.

YANG Y, NEWSAM S. Bag-of-visual-words and spatial extensions for land-use classification[C]. Proceedings of the 18th SIGSPATIAL International Conference on Advances in Geographic Information Systems. ACM , 2010: 270-279.

HUA Y, MOU L, ZHU X X. Recurrently exploring class-wise attention in a hybrid convolutional and bidirectional LSTM network for multi-label aerial image classification[J]. ISPRS journal of photogrammetry and remote sensing , 2019, 149: 188-199.

HOCHREITER S, SCHMIDHUBER J. Long short-term memory[J]. Neural computation, 1997, 9(8):1735-1780.

SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[J]. Arxiv Preprint Arxiv : 1409.1556, 2014.

Views

217

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Multi-scale dense feature fusion network for image super-resolution

Secret key extraction from atmospheric wireless optical channels by combing with generative adversarial network

Dermoscopic image classification based on multi-scale and three-dimensional interaction feature optimization

Progressive CNN-transformer semantic compensation network for polyp segmentation

Cascade residual-optimized image super-resolution reconstruction in Transformer network

Related Author

CHENG Deqiang

ZHAO Jiamin

KOU Qiqi

CHEN Liangliang

HAN Chenggong

TIAN Zhuozhan

CHEN Chunyi

HU Xiaojuan

Related Institution

School of Information and Control Engineering，China University of Mining and Technology

School of Computer Science and Technology，China University of Mining and Technology

Chongqing Research Institute， Changchun University of Science and Technology

School of Computer Science and Technology， Changchun University of Science and Technology

School of Digital and Intelligence Industry， Inner Mongolia University of Science and Technology

AI问答

Address：No.3888 Dong Nanhu Road, Changchun, Jilin, China Postal code：130033
Tel：0431-86176855 Email：gxjmgc@ciomp.ac.cn
Technical support is provided by Beijing Founder electronics co., LTD 吉ICP备11002662号-17 京公网安备11010802024621
It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.

⁰