Hexagonal Image Processing in the Context of Machine Learning: Conception of a Biologically Inspired Hexagonal Deep Learning Framework

Inspired by the human visual perception system, hexagonal image processing in the context of machine learning deals with the development of image processing systems that combine the advantages of evolutionary motivated structures based on biological models. While conventional state of the art image processing systems of recording and output devices almost exclusively utilize square arranged methods, their hexagonal counterparts offer a number of key advantages that can benefit both researchers and users. This contribution serves as a general application-oriented approach with the synthesis of the therefor designed hexagonal image processing framework, called Hexnet, the processing steps of hexagonal image transformation, and dependent methods. The results of our created test environment show that the realized framework surpasses current approaches of hexagonal image processing systems, while hexagonal artificial neural networks can benefit from the implemented hexagonal architecture. As hexagonal lattice format based deep neural networks, also called H-DNN, can be compared to their square counterpart by transforming classical square lattice based data sets into their hexagonal representation, they can also result in a reduction of trainable parameters as well as result in increased training and test rates.

[1]  Dumitru Erhan,et al.  Going deeper with convolutions , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[2]  David Middleton,et al.  Sampling and Reconstruction of Wave-Number-Limited Functions in N-Dimensional Euclidean Spaces , 1962, Inf. Control..

[3]  Rik Van de Walle,et al.  Accepted for Publication in Ieee Transactions on Image Processing Hex-splines: a Novel Spline Family for Hexagonal Lattices , 2022 .

[4]  Sameer A. Nene,et al.  Columbia Object Image Library (COIL100) , 1996 .

[5]  Hai Yang,et al.  Hexagon-Based Convolutional Neural Network for Supply-Demand Forecasting of Ride-Sourcing Services , 2019, IEEE Transactions on Intelligent Transportation Systems.

[6]  K. Goser,et al.  A low-power and high-performance CMOS fingerprint sensing and encoding architecture , 1998, Proceedings of the 24th European Solid-State Circuits Conference.

[7]  A. Hendrickson,et al.  Distribution of cones in human and monkey retina: individual variability and radial asymmetry. , 1987, Science.

[8]  Neil Storey,et al.  A Comparison Between Square and Hexagonal Sampling Methods for Pipeline Image Processing , 1990, Other Conferences.

[9]  Amos J. Storkey,et al.  School of Informatics, University of Edinburgh , 2022 .

[10]  Koray Kavukcuoglu,et al.  Exploiting Cyclic Symmetry in Convolutional Neural Networks , 2016, ICML.

[11]  Qi Feng,et al.  The analysis of VERITAS muon images using convolutional neural networks , 2016, Proceedings of the International Astronomical Union.

[12]  Bedrich J. Hosticka,et al.  Single chip CMOS image sensors for a retina implant system , 1998, ISCAS '98. Proceedings of the 1998 IEEE International Symposium on Circuits and Systems (Cat. No.98CH36187).

[13]  Tim Lukas Holch,et al.  HexagDLy - Processing hexagonally sampled data with CNNs in PyTorch , 2019, SoftwareX.

[14]  L. Gibson,et al.  Vectorization of raster images using hierarchical methods , 1982, Comput. Graph. Image Process..

[15]  Nitish Srivastava,et al.  Improving neural networks by preventing co-adaptation of feature detectors , 2012, ArXiv.

[16]  K. Sahr,et al.  Discrete Global Grid Systems , 1998 .

[17]  D. Hubel,et al.  Receptive fields and functional architecture of monkey striate cortex , 1968, The Journal of physiology.

[18]  M. Erdmann,et al.  A deep learning-based reconstruction of cosmic ray-induced air showers , 2017, 1708.00647.

[19]  Innchyn Her,et al.  Resampling on a Pseudohexagonal Grid , 1994, CVGIP Graph. Model. Image Process..

[20]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[21]  R. J. Green,et al.  Fingerprint classification using a hexagonal fast fourier transform , 1996, Pattern Recognit..

[22]  D J Field,et al.  Relations between the statistics of natural images and the response properties of cortical cells. , 1987, Journal of the Optical Society of America. A, Optics and image science.

[23]  Tom Hintz,et al.  Pseudo-invariant image transformations on a hexagonal lattice , 2000, Image Vis. Comput..

[24]  M. Trimpl,et al.  The DEPFET pixel BIOSCOPE , 1999 .

[25]  R.M. Mersereau,et al.  The processing of hexagonally sampled two-dimensional signals , 1979, Proceedings of the IEEE.

[26]  Max Welling,et al.  Group Equivariant Convolutional Networks , 2016, ICML.

[27]  T. Lohse,et al.  Application of deep learning methods to analysis of imaging atmospheric Cherenkov telescopes data , 2018, Astroparticle Physics.

[28]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[29]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[30]  Jayanthi Sivaswamy,et al.  Hexagonal Image Processing: A Practical Approach , 2014, Advances in Pattern Recognition.

[31]  Yoshua Bengio,et al.  Understanding the difficulty of training deep feedforward neural networks , 2010, AISTATS.

[32]  Innchyn Her,et al.  Geometric transformations on the hexagonal grid , 1995, IEEE Trans. Image Process..

[33]  Pierre Vandergheynst,et al.  Geometric Deep Learning: Going beyond Euclidean data , 2016, IEEE Signal Process. Mag..

[34]  Juan José Rodríguez-Vázquez,et al.  Extracting Gamma-Ray Information from Images with Convolutional Neural Network Methods on Simulated Cherenkov Telescope Array Data , 2018, ANNPR.

[35]  B.J. Hosticka,et al.  A CMOS Optical Sensor System Performing Image Sampling on a Hexagonal Grid , 1996, ESSCIRC '96: Proceedings of the 22nd European Solid-State Circuits Conference.

[36]  Harold W. Kuhn,et al.  The Hungarian method for the assignment problem , 1955, 50 Years of Integer Programming.

[37]  Eduard Gröller,et al.  Optimal regular volume sampling , 2001, Proceedings Visualization, 2001. VIS '01..

[38]  Marcel J. E. Golay,et al.  Hexagonal Parallel Pattern Transformations , 1969, IEEE Transactions on Computers.

[39]  P. Burt.  Tree and pyramid structures for coding hexagonally sampled binary images , 1980 .

[40]  Colin P.D. Birch,et al.  Rectangular and hexagonal grids used for observation, experiment and simulation in ecology , 2007 .