Fuzzy logic-based data integration: theory and applications

Abstract

Multisensor systems provide a purposeful description of the environment that a single sensor cannot offer. Because observations provided by sensors are uncertain and incomplete, we adopt the use of fuzzy sets theory as a general framework to combine uncertain measurements. We develop a fusion formula based on the measure of fuzziness. The fusion formula is mathematically tested against several desirable properties of fusion operators. We establish a fuzzification scheme by which different types of input data would be modeled. A defuzzification scheme is carried out to recover crisp data from the combined fuzzy assessment. This approach is implemented and tested with real range and intensity images acquired by an Odetics Laser Range Scanner. The goal is to obtain better scene descriptions through a segmentation process of both images. A method for evaluating segmentation results is presented.

Exploiting process integration and composition in the context of active vision

“…The most popular approaches for reasoning under uncertainty are based on probability theory [32], Dempster-Shafer theory [35], fuzzy set theory [39], and certainty factor formalism. Methods that exploit redundancy are usually based on data/sensor fusion techniques [1], [17], in which reliability is improved by pooling evidence. Sensor integration/fusion has the potential of reducing overall uncertainty, overcoming sensor imperfections, and producing more reliable results.…”

“…Sensor integration/fusion has the potential of reducing overall uncertainty, overcoming sensor imperfections, and producing more reliable results. In [1], image segmentation is performed in a framework in which fuzzy set theory is adopted for sensor fusion, where uncertainty is modeled using membership functions. A drawback of fuzzy set theory is that the estimation of the fuzzy sets or membership functions can be cumbersome if no clear guidelines can be found.…”

“…It has been shown that a moving system leads to improved robustness and the elimination of ill-posed conditions in several computer vision problems [3], [5], [6]. In addition to adjusting parameters, the visual robustness of biological systems is also due to their ability to actively integrate (fuse) 1 information from a number of visual cues [2], [29] as well as to compose behaviors in response to dynamic events.…”

Data fusion using expected output membership functions

“…We use a fuzzy set based approach because fuzzy sets can model uncertainty well. Most existing fuzzy data fusion methods use fuzzy inference techniques that use if-then rules to fuse information [5][6][7][8][9]. However, it is often difficult to choose effective if -then rules.…”

“…For example, consider a data fusion process having two inputs such as the intensity image xi and the normal image X2. Abdulghafour [9] used three fuzzy sets for each input such as I) weak edge (WE), 2) moderate edge (MO), and 3) strong edge (ST). In order to fuse inputs, he used if-then rules shown in Table 2.1.…”