Large-scale Image Recognition Method Based on Convolution Channel Selection

LI Feng; LV Yu; ZHANG Haixi; HE Guiqing

doi:10.15892/j.cnki.djzdxb.2022.02.008

Journal of Projectiles, Rockets, Missiles and Guidance >

2022 , Vol. 42 >Issue 2: 42 - 49

DOI: https://doi.org/10.15892/j.cnki.djzdxb.2022.02.008

Large-scale Image Recognition Method Based on Convolution Channel Selection

LI Feng ¹ ,
LV Yu ¹ ,
ZHANG Haixi ² ,
HE Guiqing ¹

Expand

¹ School of Electronic Information, Northwestern Polytechnical University, Xi’an 710072, China
² College of Information Engineering, Northwest Agriculture and Forestry University, Xianyang 712199, Shaanxi, China

Received date: 2021-03-16

Online published: 2025-05-29

Fold

Abstract

For a long time, due to the wide variety and different forms of large-scale images, the research and development of large-scale image recognition has been very slow. In the deep model, convolutional neural network can extract shallow features such as color and contour. With the deepening of the level, the expression of its characteristics is gradually abstracted from the characteristics of color and outline to the overall characteristics. However, we found through experiments that this feature extraction method of the network will have some "bad channels" that are not conducive to effective classification when extracting overall features. These channels participate in the subsequent calculation of the network and reduce the performance of the network to a certain extent. In order to screen out these channels that are not conducive to classification, this paper proposes a method of combining L1 and L2 norms for feature selection. By comparing the experimental results of multiple network models, the feature selection algorithm has better performance in large-scale image recognition and can improve the recognition accuracy of the network.

Key words： large-scale image recognition; convolutional neural network; feature selection; channel selection

Cite this article

LI Feng , LV Yu , ZHANG Haixi , HE Guiqing . Large-scale Image Recognition Method Based on Convolution Channel Selection[J]. Journal of Projectiles, Rockets, Missiles and Guidance, 2022 , 42(2) : 42 -49 . DOI: 10.15892/j.cnki.djzdxb.2022.02.008

[an error occurred while processing this directive]

References

Publishing order | Descend order by publishing year | Descend order by cited within

[1]	GOMES J F S, LETA F R, Applications of computer vision techniques in the agriculture and food industry: a review[J]. European Food Research and Technology, 2012, 235: 989-1000.

[2]	ZHANG S W, WANG X F, WANG Z, et al. Probability locality preserving discriminant projections for plant recognition[J]. Transactions of the Chinese Society of Agricultural Engineering, 2015, 31(11): 215-220.

[3]	DENG J, DONG W, SOCHER R, et al. Imagenet: a large-scale hierarchical image database[C]// CVPR. Proceedings of the 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. New York: IEEE, 2009: 1063-1070.

[4]	左羽, 陶倩, 吴恋, 等. 基于卷积神经网络的植物图像分类方法研究[J]. 物联网技术, 2020, 10(3):72-75.

[5]	REN S Q, HE K M, GIRSHICK R, et al. Faster R-CNN: towards real-time object detection with region proposal networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6): 1137-1149. DOI PMID

[6]	HE K M, GKIOXARI G, DOLLAR P, et al. Mask R-CNN[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020, 42(2) :386-397. DOI PMID

[7]	周飞燕, 金林鹏, 董军. 卷积神经网络研究综述[J]. 计算机学报, 2017, 40(6): 1229-1251.

[8]	卢宏涛, 张秦川. 深度卷积神经网络在计算机视觉中的应用研究综述[J]. 数据采集与处理, 2016, 31(1): 1-17.

[9]	杨末. 人脸检测及特征点精确定位方法研究[D]. 沈阳: 沈阳工业大学, 2015.

[10]	LIU X Z, TIAN Z, CHAI C Y, et al. Multiscale registration of remote sensing image using robust SIFT features in steerable-domain[J]. The Egyptian Journal of Remote Sensing and Space Science, 2011, 14(2): 63-72.

[11]	ZOU B J, GUO J J, ZHU C Z, et al. Image classification based on BOW-HOG feature[J]. Journal of Zhejiang University (Engineering Science), 2017, 51(12): 2311-2319.

[12]	LV G H, LUO S W, HUANG Y P, et al. A novel regularization method based on convolution neural network[J]. Journal of Computer Research & Development, 2014, 51 (9): 1891-1900.

[13]	ZHANG H X, HE G Q, PENG J Y, et al. Deep learning of path-based tree classifiers for large-scale plant species identification[C]// IEEE. Proceedings pf the IEEE Conference on Multimedia Information Processing and Retrieval. New York: IEEE, 2018: 25-30.

[14]	MULLER H, HERRERA A G S, KALPATHY C J, et al. Overview of the image CLEF 2012 medical image retrieval and classification tasks[EB/OL].(2012-07-12)[2021-09-20].http://citeseerx.ist.psu.edu/viewdoc/download;jsessionid=012C198C9D26251BB428F38A5C59044B?doi=10.1.1.423.2885&rep=rep1&type=pdf.

[15]	KRIZHEVSKY A, SUTSKEVER I, HINTON G E. Imagenet classification with deep convolutional neural networks[J]. Communications of the ACM, 2017, 60(6): 84-90.

[16]	SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[EB/OL].(2014-09-04)[2021-09-21]. https://arxiv.org/abs/1409.1556.

[17]	HOWARD A G, ZHU M L, CHEN B, et al. MobileNets: efficient convolutional neural networks for mobile vision applications[EB/OL]. (2017-04-17)[2021-09-21].https://doi.org/10.48550/arXiv.1704.04861.

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

Options

Outlines