Sparse arrays of signatures for online character recognition

Abstract

In mathematics the signature of a path is a collection of iterated integrals, commonly used for solving differential equations. We show that the path signature, used as a set of features for consumption by a convolutional neural network (CNN), improves the accuracy of online character recognition---that is the task of reading characters represented as a collection of paths. Using datasets of letters, numbers, Assamese and Chinese characters, we show that the first, second, and even the third iterated integrals contain useful information for consumption by a CNN. On the CASIA-OLHWDB1.1 3755 Chinese character dataset, our approach gave a test error of 3.58%, compared with 5.61% for a traditional CNN [Ciresan et al.]. A CNN trained on the CASIA-OLHWDB1.0-1.2 datasets won the ICDAR2013 Online Isolated Chinese Character recognition competition. Computationally, we have developed a sparse CNN implementation that make it practical to train CNNs with many layers of max-pooling. Extending the MNIST dataset by translations, our sparse CNN gets a test error of 0.31%.

10 pages, 2 figures

References in corpus (1)

• Improving neural networks by preventing co-adaptation of feature detectors

Cited by in corpus (29)

• Fingertip Detection and Tracking for Recognition of Air-Writing in Videos
• Recognition of Handwritten Chinese Text by Segmentation: A Segment-annotation-free Approach
• Toward high-performance online HCCR: a CNN approach with DropDistortion, path signature and spatial stochastic max-pooling
• Learning stochastic differential equations using RNN with log signature features
• Online and Offline Handwritten Chinese Character Recognition: A Comprehensive Study and New Benchmark
• Detecting early signs of depressive and manic episodes in patients with bipolar disorder using the signature-based model
• Embedding and learning with signatures
• Logsig-RNN: a novel network for robust and efficient skeleton-based action recognition
• Improved Deep Convolutional Neural Network For Online Handwritten Chinese Character Recognition using Domain-Specific Knowledge
• Character-level Chinese Writer Identification using Path Signature Feature, DropStroke and Deep CNN
• Kernels for sequentially ordered data
• Inverting the signature of a path
• Distribution Regression for Sequential Data
• Feature Engineering with Regularity Structures
• Online Signature Verification using Recurrent Neural Network and Length-normalized Path Signature
• DropSample: A New Training Method to Enhance Deep Convolutional Neural Networks for Large-Scale Unconstrained Handwritten Chinese Character Recognition
• Seq2Tens: An Efficient Representation of Sequences by Low-Rank Tensor Projections
• Signature features with the visibility transformation
• Path classification by stochastic linear recurrent neural networks
• Stroke Sequence-Dependent Deep Convolutional Neural Network for Online Handwritten Chinese Character Recognition
• Design of a Very Compact CNN Classifier for Online Handwritten Chinese Character Recognition Using DropWeight and Global Pooling
• Convergence of trapezoid rule to rough integrals
• On the signature of an image
• DeepWriterID: An End-to-end Online Text-independent Writer Identification System
• Deep Learning for Estimating Synaptic Health of Primary Neuronal Cell Culture
• DropRegion Training of Inception Font Network for High-Performance Chinese Font Recognition
• Recognition Confidence Analysis of Handwritten Chinese Character with CNN
• Learning in Sparse Rewards settings through Quality-Diversity algorithms
• A Training-based Identification Approach to VIN Adversarial Examples