Adapting Convolutional Neural Networks on the Shoeprint Retrieval for Forensic Use

Shoeprint is an important evidence for crime investigation. Many automatic shoeprint retrieval methods have been proposed in order to efficiently provide useful information for the identification of the criminals. In the mean time, the convolutional neural network shows great capacity in image classification problem but its application in shoeprint retrieval is not yet investigated. This paper presents an application of VGG16 network as feature extractor in shoeprint retrieval and a data augmentation method to fine-tune the neural network with a very small database. Our method shows a much better performance compared with state-of-the-art methods on a same database with crime-scene-like shoeprints.

[1]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[2]  Yue Li,et al.  Research of shoeprint image stream retrival algorithm with scale-invariance feature transform , 2011, 2011 International Conference on Multimedia Technology.

[3]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[4]  Madina Hamiane,et al.  A novel technique for automatic shoeprint image retrieval. , 2008, Forensic science international.

[5]  Pradeep M. Patil,et al.  Rotation and intensity invariant shoeprint matching using Gabor transform with application to forensic science , 2009, Pattern Recognit..

[6]  Jacqueline A Speir,et al.  Classification of footwear outsole patterns using Fourier transform and local interest points. , 2017, Forensic science international.

[7]  Sergio Carrato,et al.  A translational and rotational invariant descriptor for automatic footwear retrieval of real cases shoe marks , 2010, 2010 18th European Signal Processing Conference.

[8]  John Flynn,et al.  Automated processing of shoeprint images based on the Fourier transform for use in forensic science , 2005, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[9]  Pengfei Shi,et al.  Computerized Matching of Shoeprints Based on Sole Pattern , 2008, IWCF.

[10]  S. Carrato,et al.  Comparison of footwear retrieval systems for synthetic and real shoe marks , 2009, 2009 Proceedings of 6th International Symposium on Image and Signal Processing and Analysis.

[11]  Zeno Geradts,et al.  The image-database REBEZO for shoeprints with developments on automatic classification of shoe outsole designs , 1996 .

[12]  Ahmed Bouridane,et al.  Automatic Recognition of Partial Shoeprints Based on Phase-Only Correlation , 2007, 2007 IEEE International Conference on Image Processing.

[13]  Yoshua Bengio,et al.  How transferable are features in deep neural networks? , 2014, NIPS.

[14]  Rob Fergus,et al.  Visualizing and Understanding Convolutional Networks , 2013, ECCV.

[15]  Girod Alexandre,et al.  Computerized classification of the shoeprints of burglars' soles , 1996 .

[16]  Ahmed Bouridane,et al.  Automatic Recognition of Shoeprints using Fourier-Mellin Transform , 2008, 2008 NASA/ESA Conference on Adaptive Hardware and Systems.

[17]  Ahmed Bouridane,et al.  Automatic classification and recognition of shoeprints , 1999 .