图像阈值分割的Fisher准则函数法

针对图像分割中的阈值选取问题，通过引入模式识别理论中的Fisher评价函数作为图像分割的准则函数，提出了基于Fisher评价函数法的图像分割新技术。该方法利用图像直方图计算各灰度级下的Fisher评价函数值，其最大值即对应于最佳分割闲值。对该方法进行了详尽的图像分割实验，并与著名的Otsu法、最大熵法、最小误差准则法进行了详细比较，结果表明该方法具有分割性能稳定、计算速度快以及受目标大小影响小等优点，是一种实用有效的图像阈值分割新方法。

The Fisher Criterion Function Method of Image Thresholding

The threshold selection problem in image segmentation is aimed at. The Fisher evaluation function is introduced, and is acted as the criterion function of image thresholding, a new thresholding method based on Fisher criterion function is put forward. In the new method, the image gray histogram is used to compute Fisher criterion function value of all gray grades, and the optimal threshold is determined at the maximum value of Fisher criterion function. In the paper, the Fisher criterion function method is tested, and is compared with the Otsu method, the maximum entropy method, and the minimum error criterion function method. The results show that, the Fisher criterion function method has these advantages such as stable segmentation performance, fast computation speed, and very little influence by the object's size, and it is a practicable effective new image thresholding method.