Research Topics
Compression and Simplification of GPS Trajectories
Compression of GPS
Trajectories (NEW)
We consider the problem of lossy compression for
GPS trajectories with latitude, longitude and timestamp information, under a given
error tolerance, i.e., synchronous Euclidean distance.
M. Chen, M. Xu, P. Franti, "Compression of GPS Trajectories", Data Compression Conference, Snowbird, USA, 2012 (oral).
Simplifcation of GPS Trajectories
We have proposed a fast O(N)
time approximation algorithm for GPS trajectory simplification by a
joint optimization on both local integral square synchronous Euclidean distance
(LSSD) and integral square synchronous Euclidean distance (ISSD) criterion.
M. Chen, M. Xu, P. Franti, "A Fast O(N) Multi-resolution Polygonal Approximation Algorithm for GPS Trajectory Simplification", IEEE Trans. on Image Process. (Accepted)
Vector Map Compression
We have proposed algorithms for lossy
compression of vector map.
M. Chen, M. Xu and P. Fränti, "Fast
dynamic quantization algorithm for vector map compression", IEEE Int.
Conf. on Image Processing
(ICIP'10), Hong Kong, China, 4289-4292,
September 2010.
(oral)
M. Chen, M. Xu and P. Fränti,
"Optimized entropy-constrained vector quantization of lossy vector map
compression", IEEE Int. Conf. on Pattern Recognition (ICPR'10),
Istanbul, Turkey, 722-725, August 2010.
Raster Map
Image Denoising
Filtering of raster map images is chosen as a
case study of a more general class of palette-indexed images for the denoising
problem of images with a discrete number of output colors. We have proposed a
statistical filtering algorithm dealing with map images distorted by impulsive
noise, additive Gaussian noise, and mixed Gaussian-impulsive noise. The proposed
filter incorporates an information fusion process which exploits both the color
distribution in RGB space and the conditional probabilities of a given pixel in
a local context. It operates with no prior knowledge of the properties of the
noise and aims at maximal preservation of repetitive structures of the image.
M. Chen,
M. Xu, P. Fränti,
"Adaptive Context-tree based Statistical Filtering of Raster Map Images
Denoising", IEEE Trans, Multimedia,
Dec, 2011.
M. Chen,
M. Xu, P. Fränti, "Adaptive Filtering of Raster Map
Images Using Optimal Context Selection", IEEE Int. Conf. on Image Processing (ICIP’11).(oral)
M. Chen, M. Xu and P. Fränti,
"Statistical filtering of raster map images", IEEE Int. Conf. on
Multimedia & Expo (ICME'10), Singapore, 394-399, July 2010. (oral,15%)
M.
Chen,
M. Xu and P. Fränti,
"Multi-layer filtering approach for map images", IEEE Int. Conf.
on Image Processing (ICIP'09), Cairo,
Egypt, 3953-3956, November 2009.
Bit-plane
coding
We have proposed an efficient bit-plane coding
algorithm for lossless compression of gray-scale image or color palette images.
In the proposed algorithm, the context value is determined by the expectation
values of the surrounding pixels implemented by context-tree structure, in
which both the order and depth of the context template is optimized in each
bit-plane. A forgetting factor and context weighting are incorporated to
achieve a higher influence of the recent pixels.
M.
Chen,
P. Fränti and M. Xu,
"Lossless bit-plane compression of images with context tree
modeling", Int. Conf. Green Circuits and Systems (ICGCS'10),
Shanghai, China, 605-610, June 2010.
Mining Human Activities by GPS Trajectories
The focus of this work is to analyze the human
behaviour based on the collected GPS data. The collected routes are
divided into several segments with different properties (transportation modes),
such as stationary, walking, biking, running, or car driving. (PPT)