The photo-electric measurement system is a kind of high-precision measurement system for trajectory parameters and
object identity parameters, and it can acquire the image information of flying objects by CCD camera. Due to subject to
some kinds of reasons, the feature information of image is not integrated and imprecise, and it has uncertainty and fuzzy
in some degree. The Dempster-Shafer evidence theory is an important approach of uncertainty reasoning. With evidences
fused, the uncertainty of the feature information of the object is declined gradually by Dempster combination rule, so it
can achieve the aim of object detection and object recognition. The conception of fuzzy mass is expanded in the way of
the relation of absolute membership on the basis of normal mass conception. The fuzzy theory is very suitable for the
description and processing of uncertainty to evidences in D-S evidence theory, so the Basic Probability Assignment
Function (BPAF) of D-S evidence theory can be acquired according to fuzzy theory, and it resolves crucial problem in
D-S evidence theory. It is shown that data fusion method of fuzzy evidence theory can deal with uncertainty and the
fuzzy of photo-electric measurement system according to the analysis of theory and the result of experimentation, and it
has a bright future in photo-electric measurement systems.
Proc. SPIE. 6833, Electronic Imaging and Multimedia Technology V
KEYWORDS: Digital signal processing, Detection and tracking algorithms, Image segmentation, Image processing, Field programmable gate arrays, Parallel processing, Telecommunications, Signal processing, Data communications, Real time image processing
A novel parallel image processing architecture using multiple DSPs which can satisfy real-time image processing
demands is proposed, The architecture is structured with high performance DSP interconnected by FPGA. Within FPGA
the interconnection network by IRAM and the specific data communication protocol are implemented. The system
inherits merits from the tightly coupled parallel system and the loosely coupled parallel system. The system architecture
is reconfigurable and scalable. The performances measured in this platform show the high data transfer rate, and it can
satisfy parallel real-time image processing demands of the complex task, large computation and high-speed data transfer.
From the designed parallel hardware we analyze the benchmarks including acceleration ratio, parallel efficiency,
selection of processing units, interconnection network etc. Finally some suggestions are given to further improve the
system performance. The real-time image processing system based on parallel structure of multiple DSPs is easy to be
implemented. Because the system structure is reconfigurable and scalable, it is easy to change the number of DSP and
change the DSP into other series. So it has a bright future for the application of real-time image processing system.
Proc. SPIE. 6835, Infrared Materials, Devices, and Applications
KEYWORDS: Infrared search and track, Infrared sensors, Infrared imaging, Logic, Detection and tracking algorithms, Data modeling, Image processing, Kinematics, Personal digital assistants, Infrared radiation
The interacting multiple model probability data association (IMMPDA) algorithm is widely used to target tracking in
clutter. However, it is difficult for IMMPDA to get high precision track when measurements of kinematics state is inaccurate,
because it only considers kinematics feature of targets. To overcome the disadvantage, this paper presents an IMMPDA algorithm based on multi-feature fusion that utilizes multiple features of infrared targets such as kinematics state, size and gray. Association probabilities for targets position are calculated based on IMMPDA algorithm in the polar coordinates. Then the statistic distances of the size and gray are calculated according to state predictions and measurements. After that, statistic distances are further used to compute related association probabilities of targets that are in
the validation region. The decision of synthetic data association of all targets in the validation region is made based on
the information fusion, which uses fuzzy logic to get different weights of each feature. Experiments indicate that the
proposed algorithm has high quality tracking performance. Compared with conventional IMMPDA algorithm, the new algorithm cannot only get higher accurate target association but also improve the stability of the infrared target tracking system.