|
| 基于最大熵的迭代分割算法 |
| Iterative Segmentation Algorithm Based on Maximum Entropy |
| |
| DOI：10.16018/j.cnki.cn32-1650/n.202301005 |
| 中文关键词: 最大熵 图像分割 迭代算法 灰度平均值 |
| 英文关键词: maximum entropy image segmentation iterative algorithm gray average |
| 基金项目: |
|
| 摘要点击次数: 649 |
| 全文下载次数: 950 |
| 中文摘要: |
| 对于一些对象与背景像素灰度值类似的图像以及充满噪声的图像，传统的图像分割算法分割精度较低。为解决这一问题，提出了基于最大熵的迭代分割算法，根据求出的最大熵阈值将图像分为背景和对象两类区域；分别对两类区域求取灰度平均值，以该平均值将图像分为对象、背景和待分割3个区域；再对待分割区域进行迭代求取最终阈值，并根据最终阈值对图像进行分割。实验表明，该算法具有较高的抗噪性能，能精确分割一些轮廓不明显的图像，其分割精度明显好于其他传统图像分割算法。 |
| 英文摘要: |
| For some images with similar gray value of object and background pixel and images full of noise, the traditional image segmentation algorithm has low segmentation accuracy. To solve this problem, an iterative segmentation algorithm based on maximum entropy is proposed. According to the maximum entropy threshold, the image is divided into background and object regions. The gray mean value of the two types of regions is calculated, and the image is divided into three regions: object, background and to be segmented by this average. Then the final threshold is obtained iteratively for the region to be segmented, and the image is segmented according to the final threshold. Experimental results show that the algorithm has high anti-noise performance, can accurately segment some images with indistinct contour, and its segmentation accuracy is significantly better than other traditional image segmentation algorithms. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |