In this work, we propose two fast approaches (PSO and
DE) to find the optimal thresholds of an image, based on the
concept of fuzzy c-partition and maximwn entropy principle.
The image can hold as much information as possible when it is
transformed to the fuzzy domain. The algorithms are extended
to multiple thresholding. The use of PSO and DE algorithms
reduce greatly the time complexity. The two approaches were
then compared by testing them on various images. We have
found that both algorithms are comparable in terms of solution
quality when the threshold number is small. While this number
increases, PSO and DE provide the same results in terms of
accuracy and robustness of the results, but in terms of
execution time the PSO is most efficient. The experimental
results have shown the effectiveness and usefulness of the
proposed algorithms for image thresholding. The PSO and DE
algorithm can deliver satisfactory performance.