Jianqi Zhang

Learn More
This is a performance evaluation on the implementation of the Cramer-Rao lower bound (CRLB) for background clutter measurement on automatic target recognition (ATR). In essence the background clutter evaluation problem for ATR is consistent with the deterministic parameter estimation problem. Thus, useful concepts and theories of deterministic parameter(More)
The Mel frequency cepstral coefficient (MFCC) model, which is widely used in speech detection and recognition, is introduced to extract features from hyperspectral image data. The similarities and differences between speech signals and spectral image data are compared and analyzed. The standard MFCC model is then improved to suit the characteristics of(More)
Reference spectral signature selection is a fundamental work for automatic oil spill detection. To address this issue, a new approach is proposed here, which employs the density-based cluster to select a specific spectral signature from a hyperspectral image. This paper first introduces the framework of oil spill detection from hyperspectral images,(More)
A new algorithm is presented which deals with the problem of detecting small moving targets in infrared image sequences that also contain drifting and evolving clutter. Through development of models of the temporal behavior of the static background, target and cloud edge on a single pixel basis, the new algorithm employing the connecting line of the(More)
Paper documents are shredded into pieces by a shredder in what is currently a common means of ensuring text information security. Because such pieces have certain characteristics, such as being of large number and low discrimination, shredded document reconstruction by a reverse operation represents a challenge. However, recovering shredded documents is an(More)
Advances in the field of three-dimensional (3D) scanning have made the acquisition of 3D range data easier and easier. However, with the large size of 3D range data comes the challenge of storing and transmitting it. To address this challenge, this paper presents a framework to further compress 3D range data using image interpolation. We first use a virtual(More)
Display images under high illumination may appear to have lower lightness and attenuated contrast. In this paper, we review the image degradation phenomenon in ambient light through photoreceptor physiology. The influence of different environmental parameters and their effects on various gray areas in images can be evaluated by the cone response difference.(More)
The descriptive capability of color images for a scene is better than that of grayscale images. Traditional registration methods to register color images convert them to grayscale ones, in which the registration precision is decreased due to the loss of color information. By extending some properties of the discrete Fourier transform to the quaternion(More)
Detecting small target in clutter is usually carried out by using a predictor to suppress the background clutter. As an important part of clutter, noise can not be predicted accurately. Because the input value of the predicted point carries the information of noise, here a new method of background prediction algorithm is presented in this paper, which uses(More)