The oil contamination level testing is important for its using and maintenance which is the basement of the oil
contamination control is required higher by the developing device system, and the testing method is urgently needed to
be studied for improving the process method and the maintenance quality of the contaminated oil. To classify the level of
particles contamination in lubricant, CCD imaging technology is used to capture microscopic digital image of the oil
particles sample . The digital image was processed and segmented in order that the computer can recognize and
understand the particle targets by using image testing algorithm to measure the sizes, amounts and distributions of
particles. The oil contamination level can be measured effectively by the economical and convenient method in which
there is little air bubble and bead leading to false particle targets. To improve the influence produced by the false particle
targets, One method is that a series of dynamical image samples from the contaminated oil in the multi-period and the
multi-state are captured and used to test the particle targets, and the further method is to employ the fuzzy measurement
using Gaussian subjection function, which describes the distribution of the standard evidences and the distribution of the
testing data, and the testing probabilities of the evidence are weighed by the matching degree of the two distributions,
which is used to classify the oil particles contamination level .The test shows that the oil particles contamination level
diagnosis reliability is improved and the diagnosis uncertainty is reduced. This method combining with other testing
methods by using the multi-information fusion method will be further studied later.
With the rapid development of optoelectronic tracking and measurement technology, tracking equipments become more
complex and more precise, and the system faults happen at higher probability. The fault orientation, the fault analysis
and the fault exclusion change more difficult. The single information and the simple process of multi-information have
many deficiencies, which need fusion to improve the reliability. The D-S theory of evidence is a way to resolve the
uncertain problems, which fuses evidences to reason the decision results in the same recognition frame used at the
decisional level. Using the D-S theory of evidence, a diagnosis frame of multi-feature information fusion is proposed.
The deviation ranks of the fault characters is defined according to their offsets from the normal and their happening
probabilities were also computed by using the statistical results and the existing knowledge. The data reasoning of rough
set theory is employed to construct the key fault evidence space from the multi features. Further, Gaussian subjection
function from the fuzzy theory is used to describe the distribution of the key evidences and the distribution of the test
data, and the basic probabilities of the evidence are weighed by the matching degree of the two distributions. The multiperiod
and space feature information are employed and fused, and the final diagnosis decision is made by some effective
methods. A multi-feature information fusion diagnosis for the servo system of the tracking equipment is discussed. The
test shows that the diagnosis reliability is improved and the diagnosis uncertainty is reduced, and the fault diagnosis for
the precise device and other parts are also effectively resolved by using this fusion method.
With the rapid development of optoelectronic tracking and measurement technology, the testing and evaluation for the
tracking system and its inner algorithms are urgently demanded. Automobile target recognition(ATR) technology for the
image is a key part of the tracking system based on the image, and develops advanced and fast, which makes the
performance evaluation difficult and complex. There is not a reliable and effective evaluation method adaptable to the
developing technology. Therefore, a fuzzy synthesis evaluation method for ATR system or its detection and recognition
algorithms group was proposed. The evaluation indexes were selected and designed, which weights were calculated by
the direct method, W-road method and the change weight method. The simulation testing conditions, the size and the
hypothesis test methods of the statistic swatch were discussed. The mean, the covariance dependency and the distribution
indexes of the probability of detection(Pd) were effectively tested. The statistic ranges corresponding to evaluation ranks
of these indexes were established. Finally, the simple model and the division model of the fuzzy synthesis evaluation
algorithm were discussed. Tests show, this method is valuable for getting the occurrence probabilities of the different
performance racks of the system or the algorithms group corresponding to vary environment levels.
Although there are many dim target detection and tracking algorithms, they have different adaption to tasks' content and
quality. There is almost no a unified algorithm of target detection and tracking. Reliance on Automated Target
Recognition (ATR) technology is essential to future success of system reasoning and development. In labs, these
algorithms and their combination can be properly evaluated and optimized, and outfield tests and cost may be decreased.
How to analyze and evaluate these algorithms becomes an important problem of ATR. A framework of algorithms'
evaluation has been established. The parameterized simulation method of dim targets has also been proposed. This
method synthesizes simulated targets and the simulated or real background, varying the SNR(Signal Noise Ratio) and
TA(Target Area),etc, and produces quantifiable images. To evaluate the dim target detection or image pretreatment
algorithms for a single target and multiple false alarms, an effective approach SROC (Summary Receiver Operator
Characteristic) based on the ROC (Receiver Operator Characteristic) model has been improved and employed. The
dimension of FAR(False Alarm Rate) has been renewedly defined to adapt to the multiple false alarms. The SROC
model develops and quantifies the ROC model, and obtain a single performance evaluation value, which can better
quantitatively evaluate ATR algorithms. Further, the FROC(Free Receiver Operator Characteristic)model is appropriate
when multiple dim detections are possible and the number of false alarms is unconstrained. The FROC model provides a
qualified method for characterizing both the operational environment and the ability of the ATR algorithm to detect
targets. The FROC model also effectively valuates the detection performance in the situation of single dim target -
multiple false alarms. Tests show the methods are applicable and available in optimizing ATR algorithms and their
combination applications.
Some limits of standard Kalman filtering are simply analyzed. Such as the indefiniteness of motion resulted from targets
maneuvering and the lower predictive precision brought about by non or little adaptive capabilities make standard
Kalman filtering lower tracking precision and stabilization. Interacting multiple model algorithm is adopted to combine
with Kalman filter, and a new adaptive Kalman filtering algorithm for improving tracking capabilities is proposed.
Multiple models are designed to represent system possible running patterns, and "current" statistical model is designated
as one of them. Each model has an independent Kalman filter, and the general state estimation is a kind of mixing data
output produced by interacting among these models' state estimations through certain mixing probabilities. Each model
state estimation is produced by one Kalman filter corresponding this system model. In simulation tests, three system
models are designed to work, CV model , CA model and "current" statistic model. Tests show that the indefiniteness
resulted from the target motional model approximately describing the target motional pattern, the lower adaptive tracking
capabilities, and the lower tracking precision and stabilization of in targets tracking are improved efficiently. Moreover,
the strong nonlinear problem is solved effectively.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.