论文部分内容阅读
为了分割照度不均匀的网格图像,提出了一种基于最大模糊熵和遗传算法的阈值分割方法。基于模糊集合理论,根据像素灰度值把原始图像中的像素分为黑和亮两个模糊集,利用最大模糊熵准则确定模糊区间的范围,寻找模糊参数的最优组合,实现图像分割。由于穷举法搜索模糊参数的最优组合存在计算复杂度高、占用存储空间大等缺点,因此采用了遗传算法确定最优阈值。为了验证该方法的有效性,对其进行了图像分割实验,并与最大类间方差法、迭代法和一维最大熵法进行了比较。实验结果表明,该方法能够自动、有效地选取阈值,分割效果优于其它三种算法,并能保留原始图像的主要特征。
In order to segment the grid image with uneven illumination, a threshold segmentation method based on maximum fuzzy entropy and genetic algorithm is proposed. Based on the fuzzy set theory, the pixels in the original image are divided into two black and two fuzzy sets according to the pixel gray value. The maximum fuzzy entropy criterion is used to determine the range of the fuzzy interval and find the optimal combination of fuzzy parameters to achieve image segmentation. As the optimal combination of exhaustive search fuzzy parameters has the disadvantages of high computational complexity and large storage space, a genetic algorithm is used to determine the optimal threshold. In order to verify the effectiveness of the method, the image segmentation experiments were carried out, and compared with the maximum between-class variance method, iterative method and one-dimensional maximum entropy method. Experimental results show that this method can automatically and effectively select the threshold, the segmentation effect is better than the other three algorithms, and can retain the main features of the original image.