改进的二维Otsu算法在SF6压力表盘图像分割中的研究

为了提高SF6压力表盘图像分割准确度,对比了传统二维Otsu和Otsu双阈值分割算法,给出了一种改进二维Otsu算法.针对二维Otsu算法选取最佳阈值时存在的缺陷,将整体阈值再区分并结合离差平方和作为适应度函数选取最佳阈值,提高了目标图像的阈值分割效果.实验结果表明:所提算法不仅满足对SF6压力表盘图像进行目标区域分割,同时具有更高的稳定性、快速性和分割精度.

Research on improved two-dimension Otsu algorithm for SF6 pressure dial image segmentation

To enhance the image segmentation accuracy of SF6 pressure dial,an improved two-dimensional Otsu algorithm is given through comparing two threshold segmentation algorithms such as traditional two-dimensional Otsu and Otsu double threshold.Aiming at the existing defect of optimal threshold selecting by two-dimensional Otsu algorithm,the overall threshold is redistinguished and combined with the square sum of deviations which is regarded as the fitness function,which can select the best threshold value to boost threshold segmentation effect of the target image.The experimental results show that the proposed algorithm not only meets segmenting the target area on the SF6 pressure dial image,but also has higher stability,rapidity and segmentation precision.