Quantum SUSAN edge detection based on double chains quantum genetic algorithm

Abstract

Edge detection algorithm based on quantum image processing has attracted much attention due to low circuit complexity and small storage capacity. Since the classical smallest univalue segment assimilating nucleus (SUSAN) algorithm is limited in vertical and horizontal directions, a new quantum SUSAN edge detection scheme based on double chains quantum genetic algorithm is designed. First, the linear X-shift and the Y-shift transforms are raised to accelerate the qubit retrieval process and output the quantum feature information. Then, the quantum feature information is fed to quantum comparator and surround suppression circuit exports gradient information. Finally, the quantum SUSAN classifier outputs classification results. The presented algorithm combines the advantages of SUSAN’s accurate positioning and texture edge suppression with the superposition of the quantum genetic algorithm, which avoids the trap of local optimum. In the quantum SUSAN classifier circuit, the double chains quantum genetic coding and the circular SUSAN mask are introduced to improve classification accuracy. Experimental results verify that the proposed scheme has a good edge searchability. Moreover, the edge points obtained by this algorithm are more integral and continuous than other classical algorithms and existing quantum algorithms.