基于鞋印图像的性别预测方法

doi:10.19678/j.issn.1000-3428.0062940

摘要/Abstract

摘要： 鞋印是作案人在案发现场经常遗留的痕迹，承载人的性别、身高等属性信息。基于鞋印的性别预测对快速排查嫌疑人具有重要作用，其方法主要由刑侦人员凭借经验判断，需要大量领域知识，而少数自动预测方法是基于人工提取的特征和经验模型进行预测，受测量误差的影响，导致预测准确率降低。针对该问题，提出基于鞋印图像的端到端预测方法。采用卷积神经网络提取鞋印图像特征，引入通道注意力模块对特征权重进行重新分配，使模型重点关注鞋印图像中对性别起显著作用的部分。在此基础上，将特征图输入到性别预测模块进行预测。此外，分别构建适用于单枚和多枚鞋印应用场景的数据集SiSIS和SeSIS，根据在案发现场中鞋印可能出现的情况，设计鞋印方向差异、鞋印残缺和弹性形变的数据增广方式。实验结果表明，该方法在SiSIS和SeSIS数据集上的预测准确率分别达到91.80%和99.35%，相比现有基于鞋印的性别预测方法，具有较优的预测性能。

关键词: 鞋印图像, 性别预测, 卷积神经网络, 注意力机制, 生物特征识别

Abstract: Shoeprints are traces frequently left behind at crime scenes.They provide information regarding a person's attributes, such as gender and height.Gender prediction based on shoeprint plays a critical role in the rapid screening of suspects.Currently, shoeprint-based gender prediction method is mainly based on the experience of criminal investigators, which depends on extensive domain knowledge.Although a few automatic prediction methods are based on manually extracted features and empirical models, they are affected by measurement errors, reducing the prediction accuracy.This study proposes an automatic end-to-end shoeprint image-based gender prediction method.The convolutional neural network is used to extract shoeprint image features.The channel attention module is introduced to redistribute the feature weights such that the model focuses on the parts of the shoeprint image that play a significant role in gender prediction.The feature image is input into the gender prediction module for prediction using the proposed model.In addition, SiSIS and SeSIS datasets suitable for single and multiple shoeprint application scenarios are constructed.Based on possible cases of shoeprints at the crime scene, data augmentation methods of shoeprint direction difference, shoeprint imperfection, and elastic deformation are designed.The experimental results show that the prediction accuracy of the proposed method on SiSIS and SeSIS datasets are 91.80% and 99.35%, respectively.The proposed method performs better than existing gender prediction methods based on shoeprints.

Key words: shoeprint image, gender prediction, convolutional neural network, attention mechanism, biometrics recognition

中图分类号:

TP391

张涛, 朱振东, 王慧, 刘禹辰, 王新年. 基于鞋印图像的性别预测方法[J]. 计算机工程, 2022, 48(11): 306-313.

ZHANG Tao, ZHU Zhendong, WANG Hui, LIU Yuchen, WANG Xinnian. Gender Prediction Method Based on Shoeprint Image[J]. Computer Engineering, 2022, 48(11): 306-313.

http://www.ecice06.com/CN/Y2022/V48/I11/306

图/表 11

20230221182022

20230221182026

20230221182029

20230221182033

20230221182036

20230221182039

20230221182042

20230221182046

20230221182050

20230221182053

20230221182056

参考文献

[1] LU L, XU Z Y, SHI P F.Gender classification of facial images based on multiple facial regions[C]//Proceedings of WRI World Congress on Computer Science and Information Engineering.Washington D.C., USA:IEEE Press, 2009:48-52.
[2] BISWAS S, SIL J.Gender classification using spatial and temporal features[C]//Proceedings of IEEE Recent Advances in Intelligent Computational Systems.Washington D.C., USA:IEEE Press, 2013:153-157.
[3] XU Y N, ZHAO Y, ZHANG Y J.Multi-scale local binary pattern histogram for gender classification[C]//Proceedings of the 8th International Congress on Image and Signal Processing.Washington D.C., USA:IEEE Press, 2015:654-658.
[4] ARAI K, ANDRIE R.Gender classification with human gait based on skeleton model[C]//Proceedings of the 10th International Conference on Information Technology.Washington D.C., USA:IEEE Press, 2013:113-118.
[5] HU M D, WANG Y H.A new approach for gender classification based on gait analysis[C]//Proceedings of the 5th International Conference on Image and Graphics.Washington D.C., USA:IEEE Press, 2009:869-874.
[6] HEMA M, ESTHER R K.Gait energy image projections based on gender detection using support vector machines[C]//Proceedings of the 5th International Conference on Communication and Electronics Systems.Washington D.C., USA:IEEE Press, 2020:1315-1320.
[7] BODZIAK W J.Footwear impression evidence:detection, recovery and examination[M].2nd ed.Boca Raton, USA:CRC Press, 2000:383-385.
[8] 史力民.足迹检验实用技术教程[M].北京:中国人民公安大学出版社, 2012. SHI L M.Practical technology course of footprint inspection[M].Beijing:People's Public Security University of China Press, 2012.(in Chinese)
[9] ZEYBEK G, ERGUR I, DEMIROGLU Z.Stature and gender estimation using foot measurements[J].Forensic Science International, 2008, 181(1/2/3):54.
[10] OZDEN H, BALCI Y, DEMIRÜSTÜ C, et al.Stature and sex estimate using foot and shoe dimensions[J].Forensic Science International, 2005, 147(2/3):181-184.
[11] ATAMTURK D.Estimation of sex from the dimensions of foot, footprints, and shoe[J].Anthropologischer Anzeiger, 2010, 68(1):21-29.
[12] ABLEDU J K, ABLEDU G K, OFFEI E B, et al.Determination of sex from footprint dimensions in a Ghanaian population[J].PLoS One, 2015, 10(10):1-13.
[13] HEMY N, FLAVEL A, ISHAK N I, et al.Sex estimation using anthropometry of feet and footprints in a western Australian population[J].Forensic Science International, 2013, 231(1/2/3):402.
[14] SEN J, KANCHAN T, GHOSH S.Sex estimation from foot dimensions in an indigenous Indian population[J].Journal of Forensic Sciences, 2011, 56(Suppl 1):148-153.
[15] DAYARATHNE S, NAWARATHNA L S, NANAYAKKARA D.Determination gender using foot, footprint, hand and hand print measurements in a Sinhalese population in Sri Lanka using supervised learning techniques[J].Computer Methods and Programs in Biomedicine Update, 2021, 1:1-17.
[16] CHOOSAKOONKRIANG S, KHEAWPUM O.Estimation of stature and sex from step length and shoe dimensions for forensic investigation[J].Interdisciplinary Research Review, 2020, 15(6):1-5.
[17] 史力民, 李硕, 赵悦岑.基于深度学习的赤足迹性别自动分析研究[J].中国刑警学院学报, 2018(3):97-99. SHI L M, LI S, ZHAO Y C.Study on the gender automatic analysis of red footprint based on deep learning[J].Journal of Criminal Investigation Police University of China, 2018(3):97-99.(in Chinese)
[18] 王新年, 于丹, 张涛.穿鞋足迹序列的足迹能量图组表达与识别[J].中国图象图形学报, 2021, 26(10):2357-2375. WANG X N, YU D, ZHANG T.Shoeprint sequence representation and recognition using shoeprint energy map set[J].Journal of Image and Graphics, 2021, 26(10):2357-2375.(in Chinese)
[19] SZEGEDY C, VANHOUCKE V, IOFFE S, et al.Rethinking the inception architecture for computer vision[C]//Proceedings of Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2016:2818-2826.
[20] CHEN Q, LIU L, HAN R, et al.Image identification method on high speed railway contact network based on YOLO v3 and SENet[C]//Proceedings of Chinese Control Conference.Washington D.C., USA:IEEE Press, 2019:8772-8777.
[21] LV S Q, WANG Z P, NA J Y.Arrhythmia classification of merged features method based on SENet and BiLSTM[C]//Proceedings of the 4th International Conference on Information Communication and Signal Processing.Washington D.C., USA:IEEE Press, 2021:162-167.
[22] LU J S, XIONG C M, PARIKH D, et al.Knowing when to look:adaptive attention via a visual sentinel for image captioning[C]//Proceedings of Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2017:3242-3250.
[23] WU M, JIN X, JIANG Q, et al.Remote sensing image colorization based on multiscale SENet GAN[C]//Proceedings of the 12th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics.Washington D.C., USA:IEEE Press, 2019:1-6.
[24] HU J, SHEN L, ALBANIE S, et al.Squeeze-and-excitation networks[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020, 42(8):2011-2023.
[25] MEHTA S, PAUNWALA C, VAIDYA B.CNN based traffic sign classification using adam optimizer[C]//Proceedings of International Conference on Intelligent Computing and Control Systems.Washington D.C., USA:IEEE Press, 2019:1293-1298.
[26] WANG H.Garbage recognition and classification system based on convolutional neural network VGG16[C]//Proceedings of the 3rd International Conference on Advanced Electronic Materials, Computers and Software Engineering.Washington D.C., USA:IEEE Press, 2020:252-255.

选择文件类型/文献管理软件名称

选择包含的内容