Corpus ID: 10261024

Detecting Good Quality Frames in Videos from Mobile Camera for Blind Navigation

  title={Detecting Good Quality Frames in Videos from Mobile Camera for Blind Navigation},
  author={Long Tian and Chucai Yi and Yingli Tian}
—The development of smart mobile device with cameras makes it possible to build wearable and portable blind assistant navigation systems. However, it is difficult for blind user to capture high-quality images and videos of surrounding environments without motion blur or de-focus blur. To avoid this blur effect in camera-based videos, this paper proposes a method of high-quality frame detection based on image quality assessment. It distinguishes blurred frames from unblurred frames in the videos… Expand

Figures and Tables from this paper

Recognizing Elevator Buttons and Labels for Blind Navigation
  • Jingya Liu, Yingli Tian
  • Computer Science
  • 2017 IEEE 7th Annual International Conference on CYBER Technology in Automation, Control, and Intelligent Systems (CYBER)
  • 2017
Preliminary results demonstrate the robustness and effectiveness of the proposed method for elevator button detection and associated label recognition. Expand
Learning Alphabets and Numbers from Hand Motion Trajectories for Visually Impaired Children
A computer vision-based system to recognize alphabets and numbers from hand motion trajectories captured by an RGB-D camera that achieves 86% accuracy for subject independent test. Expand


Detecting good quality frames in videos captured by a wearable camera for blind navigation
A new method to detect good quality frames from videos captured by cameras, which are taken by blind users, by using a support vector machine based classifier to identify the frames with good quality (Unblurred) from those blurred frames. Expand
Detecting and Removing the Motion Blurring from Video Clips
A framework to deblur the blurry frame in a video clip to remove the blurring effect caused by hand shaking and the background is reconstructed by the algorithm in each frame, so an accurate matte of blurry object can be extracted to debLur the moving object by alpha matting. Expand
Learning without Human Scores for Blind Image Quality Assessment
The proposed QAC based BIQA method not only has comparable accuracy to those methods using human scored images in learning, but also has merits such as high linearity to human perception of image quality, real-time implementation and availability of image local quality map. Expand
Robust and Effective Component-Based Banknote Recognition for the Blind
A novel camera-based computer vision technology to automatically recognize banknotes to assist visually impaired people and employs the spatial relationship of matched SURF features to detect if there is a bill in the camera view, which largely alleviates false recognition. Expand
Blind Image Quality Assessment: A Natural Scene Statistics Approach in the DCT Domain
An efficient general-purpose blind/no-reference image quality assessment (IQA) algorithm using a natural scene statistics model of discrete cosine transform (DCT) coefficients, which requires minimal training and adopts a simple probabilistic model for score prediction. Expand
Discriminative Blur Detection Features
This work addresses the question of what are effective features to differentiate between blurred and unblurred image regions by studying a few blur feature representations in image gradient, Fourier domain, and data-driven local filters. Expand
Assistive Text Reading from Complex Background for Blind Persons
A camera-based assistive system for visually impaired or blind persons to read text from signage and objects that are held in the hand and then communicate this information aurally, which outperforms previous algorithms on some measures. Expand
No-Reference Image Quality Assessment in the Spatial Domain
Despite its simplicity, it is able to show that BRISQUE is statistically better than the full-reference peak signal-to-noise ratio and the structural similarity index, and is highly competitive with respect to all present-day distortion-generic NR IQA algorithms. Expand
Automatic detection and recognition of signs from natural scenes
This paper proposes a local intensity normalization method to effectively handle lighting variations, followed by a Gabor transform to obtain local features, and finally a linear discriminant analysis (LDA) method for feature selection. Expand
No-reference image quality assessment using modified extreme learning machine classifier
The experimental results prove that the estimated visual quality of the proposed RCGA-ELM emulates the mean opinion score very well and indicate that the proposed schemes significantly improve the performance of ELM classifier under imbalance condition for image quality assessment. Expand