图像阈值分割算法及对比研究

Image Thresholding Segmentation Algorithm and Contrastive Study

图像阈值分割技术广泛应用于计算机视觉和模式识别等研究领域。通过对五种典型的阈值分割算法,包括最大类间方差法、最大熵法、最小交叉熵法、最大相关法和灰度熵法进行算法仿真,并针对不同信噪比图像的分割结果进行了客观评价,得出不同分割方法所适用的分割对象。同时设计了图像阈值分割系统,便于进行不同分割算法的对比分析。

Image thresholding segmentation is widely used in the fields of computer vision and pattern recognition. Five classi-cal thresholding methods have been simulated, including maximum interclass variance method, maximum entropy method, minimum cross entropy method, maximum correlation method and gray entropy method.The segmentation results are quantita-tively evaluated by some images with different SNR. The conclusion that objects to be segmented are adapted to each algorithm has been drawn. Meanwhile, the image thresholding segmentation system is designed to analyze the different segmentation al-gorithms.