New Fusion Based Enhancement for Text Detection in Night Video Footage

Text Detection in night video footage is hard due to low contrast and low resolution caused by distance variations between camera and ground under poor light. In this paper, we propose a new fusion based enhancement method for text detection especially in night video footage. The proposed method integrates the merits of color space and frequency based enhanced methods for sharpening low contrast details. Specifically, for each enhanced image, the proposed method derives weighted mean for the pixels values to widen the gap between high contrast (texts) and low contrast (background) pixels. The weighed means are further modified as dynamic weights with respect to enhanced images. These weights are convolved with pixel values of respective enhanced images to produce fused images. The proposed fusion based enhancement method is tested on images collected from night video footage to demonstrate the effectiveness of the method. For the output of each enhancement method including the proposed method, text detection rates are computed to show that the proposed enhancement method outperforms the existing enhancement methods.

[1]  Youren Wang,et al.  Medical image fusion using discrete fractional wavelet transform , 2016, Biomed. Signal Process. Control..

[2]  Moon Gi Kang,et al.  Edge enhancement algorithm for low-dose X-ray fluoroscopic imaging , 2017, Comput. Methods Programs Biomed..

[3]  Xu-Cheng Yin,et al.  Text Detection, Tracking and Recognition in Video: A Comprehensive Survey , 2016, IEEE Transactions on Image Processing.

[4]  Amod Kumar,et al.  A social spider optimized image fusion approach for contrast enhancement and brightness preservation , 2017, Appl. Soft Comput..

[5]  David S. Doermann,et al.  Text Detection and Recognition in Imagery: A Survey , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6]  Shuchang Zhou,et al.  EAST: An Efficient and Accurate Scene Text Detector , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[7]  Wang Rui,et al.  Medical X-ray image enhancement method based on TV-homomorphic filter , 2017, 2017 2nd International Conference on Image, Vision and Computing (ICIVC).

[8]  Pan He,et al.  Detecting Text in Natural Image with Connectionist Text Proposal Network , 2016, ECCV.

[9]  Ernest Valveny,et al.  ICDAR 2015 competition on Robust Reading , 2015, 2015 13th International Conference on Document Analysis and Recognition (ICDAR).

[10]  Ya Su,et al.  A Unified Framework for Tracking Based Text Detection and Recognition from Web Videos , 2018, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[11]  Tong Lu,et al.  Fractional poisson enhancement model for text detection and recognition in video frames , 2016, Pattern Recognit..

[12]  Xiang Bai,et al.  Detecting Oriented Text in Natural Images by Linking Segments , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[13]  Ju Jia Zou,et al.  Contrast Enhancement Using Region Based Image Fusion in Wavelet Domain , 2016, Robotics and Mechatronics.

[14]  Wei Zeng,et al.  Night video enhancement using improved dark channel prior , 2013, 2013 IEEE International Conference on Image Processing.

[15]  Chun Yang,et al.  Multi-orientation scene text detection with multi-information fusion , 2016, 2016 23rd International Conference on Pattern Recognition (ICPR).