New Approach to Image Edge Detection Based on Quantum Entropy

Abstract

We propose a novel edge detection algorithm based on quantum entropy using a flexible representation of the quantum image. We use information entropy to measure the amount of information contained in digital images in view of quantum information processing. Quantum entropy can take correlations among quantum bases into the calculation of entropy, while Shannon entropy is powerless on this, namely, quantum entropy is more accurate than Shannon entropy in quantum information measurements. Therefore, the quasithreshold that leads to maximum quantum entropy should be adopted as the optimal threshold, because the maximum amount of information is obtained under this circumstance. The quantum version of the image segmentation works with computational basis states, exclusively. We prove the efficiency of the approach proposed on examples from the real world, microscopy, microarray, medical, and satellite images. We present the performance evaluation of the proposed technique based on the peak-signal-to-noise ratio.