Chengzhang Qu

Learn More
Password-based authentication is easy to use but its security is bounded by how much a user can remember. Biometrics-based authentication requires no memorization but ‘resetting’ a biometric password may not always be possible. In this paper, we propose a user-friendly authentication system (KinWrite) that allows users to choose arbitrary, short and(More)
A novel approach is proposed in this paper for automatic forest fire detection from video. Based on 3D point cloud of the collected sample fire pixels, Gaussian mixture model is built and helps segment some possible flame regions in single image. Then the new specific flame pattern is defined for forest, and three types of fire colors are labeled(More)
In this paper, we propose a real-time forest fire detection algorithm using artificial neural networks based on dynamic characteristics of fire regions segmented from video images. Fire region is obtained from image with the help of threshold values in HSV color space. Area, roundness and contour are computed for fire regions from each 5 continuous frames.(More)
In the methods of video based fire detection, wavelet was usually used to test whether a pixel belongs to fire region, while fast Fourier transform was used to describe the contour of fire area. We try these two methods on images with forest flame, and develop one new approach combined both wavelet and FFT. In our method, contour of fire is found firstly,(More)
In material science and engineering, the grain structure inside a super-alloy sample determines its mechanical and physical properties. In this paper, we develop a new Multichannel Edge-Weighted Centroidal Voronoi Tessellation (MCEWCVT) algorithm to automatically segment all the 3D grains from microscopic images of a super-alloy sample. Built upon the(More)
The existing approaches for object detection from remote sensing images usually have the assumptions that the location is already known or determined manually. Our paper proposes an automatic and rapid method to detect objects from satellite image with large size, which is the precondition for detailed object recognition. Since feature based method usually(More)
This paper proposes a new color-based method for forest fire segmentation from video image. Two kinds of color features including intensity value and color distribution are used. In segmentation based on V value of HSV color model, the lower threshold is defined from experience, and the higher threshold value is computed automatically, thus the possible(More)
This paper proposes novel algorithms for color based segmentation and shape based matching of forest fires from video frames. Fire pixels from the sample images are collected and compose 3D point cloud in RGB color space, based on which Gaussian mixture model is used to build a more precise color model. According to new definition of flame pattern, possible(More)
Detecting human interactions in a public place where no physical touch occurs has important applications in many surveillance tasks. In this paper, we explore the possibilities to detect such distant human interactions without recognizing the specific human actions. Specifically, we use a simplified formulation of the interaction in this paper: when two(More)
This paper proposes one method for the extraction of eye contour from monocular image with monkey face. Face region is segmented using the facial skin color model constructed from some samples and the operations of mathematical morphology. Then the face region is size normalized, and rotation normalized based on the located eyes. Taking the boundary of eye(More)