基于灰度-梯度二维对称Tsallis交叉熵的阈值分割

Thresholding segmentation based on gray-gradient 2-D symmetric Tsallis cross entropy

针对灰度级-平均灰度级直方图的二维Tsallis交叉熵阈值分割法存在错分、计算复杂度较高问题,提出一种基于灰度-梯度二维对称Tsallis交叉熵的阈值分割方法。构建新的灰度-梯度二维直方图,更加全面地考虑目标点和背景点;导出基于该直方图区域划分的对称Tsallis交叉熵阈值选取公式;采用基于tent映射的混沌小生境粒子群优化算法搜寻二维最佳阈值向量,并引入快速递推算法降低其适应度函数的计算复杂度。实验结果表明,与基于灰度级-平均灰度级直方图的二维Tsallis交叉熵阈值分割法相比,该方法能够使分割后的图像边缘更加准确,类内灰度更加均匀,且实时性提高了30倍。

2-D Tsallis cross entropy thresholding segmentation based on gray level-average gray level histogram exists misclassification, higher computational complexity. So thresholding segmentation based on gray-gradient 2-D symmetric Tsallis cross entropy is proposed. Firstly, a new 2-D histogram based on gray-gradient is created, and more fully considers the object points and background points. Then, the formulas of symmetric Tsallis cross entropy threshold selection based on the histogram zoning are derived. Finally, chaotic niche particle swarm optimization algorithm based on tent map is used to search for 2-D optimal threshold vector, and fast recursive algorithm is introduced to reduce the computational complexity of the fitness function. Experimental results show that this method enables the edge of the segmented image more accurate, more uniform gray within the class, and real-time increased by 30-fold, compared with 2-D Tsallis cross entropy segmentation based on gray level-average gray level histogram.