Robust Scene Text Recognition Using Sparse Coding based Features

TL;DR

This paper introduces a novel scene text recognition approach using sparse coding based features called HSC, which improves detection accuracy over traditional HOG features by leveraging learned dictionaries and integrates multiple cues for robust word recognition.

Contribution

The paper presents a new sparse coding based feature (HSC) for scene text recognition, replacing HOG features, and demonstrates its effectiveness through comprehensive experiments.

Findings

HSC features outperform HOG features in character detection.

The method achieves superior accuracy on challenging datasets.

Integration of multiple cues enhances recognition robustness.

Abstract

In this paper, we propose an effective scene text recognition method using sparse coding based features, called Histograms of Sparse Codes (HSC) features. For character detection, we use the HSC features instead of using the Histograms of Oriented Gradients (HOG) features. The HSC features are extracted by computing sparse codes with dictionaries that are learned from data using K-SVD, and aggregating per-pixel sparse codes to form local histograms. For word recognition, we integrate multiple cues including character detection scores and geometric contexts in an objective function. The final recognition results are obtained by searching for the words which correspond to the maximum value of the objective function. The parameters in the objective function are learned using the Minimum Classification Error (MCE) training method. Experiments on several challenging datasets demonstrate that the proposed HSC-based scene text recognition method outperforms HOG-based methods significantly and outperforms most state-of-the-art methods.