Low-complexity DQED: Advancing dual-scenario quantum edge detection for enhanced image analysis

To address the existing problems of complex process, including complex pixel operations, high complexity and cost, and single scenario of existing quantum edge detection, we propose a low-complexity Dual-Scenario Quantum Image Edge Detection (DQED) method which applies for dual scenarios: Contour Edge Detection (CED) for coarse edge detection and Texture Edge Detection (TED) for detail edge detection. In DQED, edge information is detected using only one Controlled-Controlled-NOT gate (CCNOT) gate without complex operations. To simplify the detection process, we propose the Neighborhood Quantum State-based Edge Extraction (NQEE) method, which uses only the binary image of the object image and the Highest Weight Qubit (HWQ) plane to detect the edge. Moreover, to reduce the complexity, we discard the complex pixel-based operations by using only XOR operations in the NQEE. In addition, to refine the edge image, we propose the Quantum Edge Refinement (QER) algorithm, which is used in both the CED and TED processes to obtain the contour edge and the texture edge. This paper clearly describes the proposed methods and designs the quantum circuits in detail. Finally, we fully evaluate our method with images from seven databases that are of different characteristics. We also consider quantum channel noise and evaluate it. Comparison with the existing state-of-the-art research results show that our method has the advantages of generalization, dual scenarios, simplicity, and low complexity.