This paper presents a method for the detection of small objects from the infrared images. The detection is performed on the intensity surface well fitted by the cubic facet model. The small target energy distribution ...
详细信息
This paper presents a method for the detection of small objects from the infrared images. The detection is performed on the intensity surface well fitted by the cubic facet model. The small target energy distribution presents as a convex surface on the image intensity surface and the target center is the maximal extremum points of the convex surface. According to the extremum theory, the possible small target position is analytically determined by directly convolving the original image with the derivative operators deduced from the bivariate cubic function. With the available coarse target locations, the potential target is separated from the background by examining the intensity features of the target cluster. Experimental results on the sample infrared images demonstrate the proposed algorithm provides a robust and efficient performance.
A noise erosion operator based on partial differential equation (PDE) is introduced, which has an excellent ability of noise removal and edge preservation for two-dimensional (2D) gradient data. The operator is applie...
详细信息
A noise erosion operator based on partial differential equation (PDE) is introduced, which has an excellent ability of noise removal and edge preservation for two-dimensional (2D) gradient data. The operator is applied to estimate a new diffusion coefficient. Experimental results demonstrate that anisotropic diffusion based on this new erosion operator can efficiently reduce noise and sharpen object boundaries.
We propose an adaptive model update mechanism for mean-shift blob tracking. It is novel for us to use self-tuning Kalman filters for estimating object kernel-color distribution, i.e. kernel-histogram. Filtering residu...
详细信息
Motivated by ideas of group representation theory, we propose a matrix-oriented method to dimension reduction for image data. By virtue of the action of Stiefel manifold, the original image representations can be dire...
详细信息
In this paper, we propose a new automated approach to extract the centerlines from 2-D angiography. The centerline extraction is the basis of 3-D reconstruction of the blood vessels, so the accurate localization of ce...
详细信息
Thermal swipe sensor is one of the technologies behind many of today's fingerprint sensors. This paper presents a new direct Fourier-based algorithm for performing swipe fingerprints registration to subpixel accur...
详细信息
A new technology evaluation of fingerprint verification algorithms has been organized following the approach of the previous FVC2000 and FVC2002 evaluations, with the aim of tracking the quickly evolving state-ofthe- ...
详细信息
The residue number system (RNS) has computational advantages in addition and multiplication compared with weighted number systems, such as the binary number system (BNS), since operations on residue digits are perform...
详细信息
The residue number system (RNS) has computational advantages in addition and multiplication compared with weighted number systems, such as the binary number system (BNS), since operations on residue digits are performed independently and these processes can be performed in parallel. Thus they are widely used in digital signal processing etc. Since residue to binary conversion is critical and difficult for the practicality of RNS, in this paper, a novel residue to binary (R/B) conversion algorithm for the restricted moduli set (2/sup n/ -1, 2/sup n/, 2n+1), based on exploring the periodicity of modulo (2/sup n/ /spl plusmn/ 1) operations is presented. A new 2n-bit adder based R/B converter is also proposed. The performance comparison results demonstrate that the new converter is faster and requires less area compared with the others reported in the previous literature.
Spatial data mining refers to extracting and "mining" the hidden, implicit, valid, novel and interesting spatial or non-spatial patterns or rules from large-amount, incomplete, noisy, fuzzy, random, and prac...
详细信息
ISBN:
(纸本)0780384032
Spatial data mining refers to extracting and "mining" the hidden, implicit, valid, novel and interesting spatial or non-spatial patterns or rules from large-amount, incomplete, noisy, fuzzy, random, and practical spatial databases. In which an important issue but remains underdeveloped is to reveal and handle the uncertainties in spatial data mining. In This work, uncertainty of spatial data is briefly analyzed firstly, including the types and origins of uncertainty, their models of measurement and propagation. Then, some uncertainty factors in operation of spatial data mining are discussed and some uncertainty handling methods are adopted, including maximum variance data discretization and fuzzy belief function. Finally, we think the process of spatial data mining can be regarded as a complex system, a linear serial processing system in engineering control systems. An uncertainty propagation model of spatial data mining - fuzzy logic uncertainty propagation model with credibility factor is developed. Moreover, several key problems about uncertainty handling and propagation in spatial data mining are put forward.
暂无评论