Classification of engraved pottery sherds mixing deep-learning features by compact bilinear pooling

Abstract The ARCADIA project aims at using pattern recognition and machine learning to promote a systematic analysis of the large corpus of archaeological pottery fragments excavated in Saran (France). Dating from the High Middle Ages, these sherds have been engraved with repeated patterns using a carved wooden wheel. The study of these engraved patterns allows archaeologists to better understand the diffusion of ceramic productions. In this paper, we present a method that classifies patterns of ceramic sherds by combining deep learning-based features extracted from some pre-trained Convolutional Neural Network (CNN) models. A dataset composed of 888 digital patterns extracted from 3D scans of pottery sherds was used to evaluate our approach. The classification capacity of each CNN model was first assessed individually. Then, several combinations of common pooling methods using different classifiers were tested. The best result was obtained when features of the VGG19 and ResNet50 models were combined using Compact Bilinear Pooling (CBP) with a high classification rate of 95.23%.

[1]  Aladine Chetouani,et al.  Classification of Ceramic Shards Based on Convolutional Neural Network , 2018, 2018 25th IEEE International Conference on Image Processing (ICIP).

[2]  Fernando Zvietcovich,et al.  A novel method for estimating the complete 3D shape of pottery with axial symmetry from single potsherds based on principal component analysis , 2016, Digit. Appl. Archaeol. Cult. Heritage.

[3]  Jianbo Li,et al.  Reassembling 3D Thin Fragments of Unknown Geometry in Cultural Heritage , 2014 .

[4]  Yang Gao,et al.  Compact Bilinear Pooling , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[5]  Filippo Stanco,et al.  Computer graphics solutions for pottery colors specification , 2013, Electronic Imaging.

[6]  Jun Zhou,et al.  Identifying Designs from Incomplete, Fragmented Cultural Heritage Objects by Curve-Pattern Matching , 2016, J. Electronic Imaging.

[7]  Song Wang,et al.  Design Identification of Curve Patterns on Cultural Heritage Objects: Combining Template Matching and CNN-based Re-Ranking , 2018, ArXiv.

[8]  Aladine Chetouani,et al.  Automatic classification of ceramic sherds with relief motifs , 2017, J. Electronic Imaging.

[9]  David B. Cooper,et al.  Accurately Estimating Sherd 3D Surface Geometry with Application to Pot Reconstruction , 2003, 2003 Conference on Computer Vision and Pattern Recognition Workshop.

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

[11]  Ali Shokoufandeh,et al.  Classification of archaeological ceramic fragments using texture and color descriptors , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Workshops.

[12]  Rasmus Pagh,et al.  Fast and scalable polynomial kernels via explicit feature maps , 2013, KDD.

[13]  Martin Kampel,et al.  Color classification of archaeological fragments , 2000, Proceedings 15th International Conference on Pattern Recognition. ICPR-2000.

[14]  Jean Ponce,et al.  A Theoretical Analysis of Feature Pooling in Visual Recognition , 2010, ICML.

[15]  Martin Kampel,et al.  Rule based system for archaeological pottery classification , 2007, Pattern Recognit. Lett..

[16]  Petros Daras,et al.  Automatic classification of archaeological pottery sherds , 2012, JOCCH.

[17]  Luciano Silva,et al.  3D reconstruction methods for digital preservation of cultural heritage: A survey , 2014, Pattern Recognit. Lett..

[18]  Majdi Khoudeir,et al.  Gabor Filter-Based Texture Features to Archaeological Ceramic Materials Characterization , 2012, ICISP.

[19]  Subhransu Maji,et al.  Bilinear CNN Models for Fine-Grained Visual Recognition , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[20]  Hans-Peter Kriegel,et al.  A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise , 1996, KDD.

[21]  Luca Di Angelo,et al.  Axis estimation of thin-walled axially symmetric solids , 2018, Pattern Recognit. Lett..

[22]  Wageeh Boles,et al.  Multispectral Periocular Classification With Multimodal Compact Multi-Linear Pooling , 2017, IEEE Access.

[23]  David B. Cooper,et al.  Axially Symmetric 3D Pots Configuration System Using Axis of Symmetry and Break Curve , 2013, 2013 IEEE Conference on Computer Vision and Pattern Recognition.

[24]  Tom Drummond,et al.  Machine Learning for High-Speed Corner Detection , 2006, ECCV.

[25]  Alex Krizhevsky,et al.  One weird trick for parallelizing convolutional neural networks , 2014, ArXiv.