Tracking Based Multi-Orientation Scene Text Detection: A Unified Framework with Dynamic Programming

Abstract

There are a variety of grand challenges for multi orientation text detection in scene videos, where the typical issues include skew distortion, low contrast and arbitrary motion. Most conventional video text detection methods using individual frames have limited performance. In this paper, we propose a novel tracking based multi-orientation scene text detection method using multiple frames within a unified framework via dynamic programming. First, a multi-information fusion based multi orientation text detection method in each frame is proposed to extensively locate possible character candidates and extract text regions with multiple channels and scales. Second, an optimal tracking trajectory is learned and linked globally over consecutive frames by dynamic programming to finally refine the detection results with all detection, recognition and prediction information.

Existing System

Neural Network, Adaptive Clustering.

Proposed System

In this paper, we construct a tracking based text detection system in scene videos by locating character candidates extensively and searching text regions globally. In our approach, the multi-information fusion strategy can precisely detect multi-orientation text regions with fair recall by extensively extracting character candidates from multiple channels and scales and powerfully filtering non-text region candidates. The tracking based text detection technique can robustly detect video text with high average tracking accuracy by tracking text with multiple tracking strategies and integrating detection results with dynamic programming.

Architecture

BLOCK DIAGRAM