侧扫声呐图像分割的中性集合与量子粒子群算法

The Neutrosophic Set and Quantum-behaved Particle Swarm Optimization Algorithm of Side Scan Sonar Image Segmentation

针对现有的侧扫声呐图像分割方法存在分割准确率不高和效率偏低的问题,提出了一种基于中性集合和量子粒子群算法的侧扫声呐图像阈值分割方法。通过基于中性集合计算图像灰度共生矩阵,实现了侧扫声呐图像精细纹理的表达,提高了分割精度;基于二维最大熵理论,采用量子粒子群算法计算二维最优分割阈值向量,实现了分割阈值向量的快速准确获取,提高了分割效率和精度。最终实现了高噪声侧扫声呐图像目标的准确、高效分割。通过对含有不同目标的侧扫声呐图像的分割试验,验证了该算法的有效性。

Due to the problem of the existing image segmentation methods applied in side scan sonar （SSS）image often suffered from low efficiency or low accuracy,this paper proposed a novel SSS image thresholding segmentation method based on neutrosophic set （NS）and quantum-behaved particle swarm optimization （QPSO）algorithm.Firstly,the image gray co-occurrence matrix is constructed in NS domain, the fine texture of SSS image is expressed,and this can improve the accuracy of SSS image segmentation. Then,based on the two-dimensional maximum entropy theory,the optimal two-dimensional segmentation threshold vector is quickly and accurately obtained by QPSO algorithm,and this can improve the efficiency and accuracy of SSS image segmentation.Finally,the accurate and high efficient target segmentation of SSS image with high noises is realized.The effectiveness of the algorithm is verified by segmenting SSS image containing different targets.