Yingkui Du

  • Citations Per Year
Learn More
Shadows are physical phenomena observed in most natural scenes. They can cause many problems in computer vision performance. The paper addresses the problem of shadow detection and removal from solo image of natural scenes. Our method is based on Retinex theory which is an image enhancement and illumination compensation model of the lightness and color(More)
The Hough Transform (HT), the Radon Transform (RT) and the Line Segment Detector (LSD) are the most well-known methods for line detection. But the HT and RT methods cost large computing consumption and always resulting a poor performance with many outliers. The LSD method is not effective to the real application of complex background condition. In this(More)
In this paper, a supervised approach to online learn a structured sparse and discriminative representation for object tracking is presented. Label information from training data is incorporated into the dictionary learning process to construct a compact and discriminative dictionary. This is accomplished by adding an ideal-code regularization term and(More)
A robust progressive structure from motion (PSFM) method is proposed for unordered images. Our method can reduce accumulative error efficiently during scene dense recovery and camera motion estimation. The whole unordered images are divided into two classes: key frames and non-key frames. For key frames, superior features are selected and tracked to(More)
Noise removal is an important image processing problem which has wide applications in different fields. The noise removal and edge preservation algorithm based on variational method and partial differential equation (PDE) for processing Magnetic Resonance images is proposed. The proposed method can avoid the blocky effects and false edges, compared with(More)
Automatic power line recognition from cluttered background is an important and challenging task for a vision based unmanned aerial vehicles (UAVs) power line inspection system. In this paper, we propose a power line recognition method based on liner object enhancement and parallel lines constraint. A new double-side filter is proposed to enhance the power(More)