|
| 摘要: |
| 分水岭分割方法虽然是一种有效且常用的图像分割方法,但它容易产生过分割现象且对噪声敏感。针对这一问题,本文结合小波变换提出了一种基于梯度校正的分水岭图像分割方法。首先利用小波阈值去噪,然后应用最大熵算法对梯度图像进行校正,最后基于校正过的梯度图像进行图像分割。数值实验结果表明,本文提出的分割算法取得了良好的图像分割效果。 |
| 关键词: 图像分割 分水岭算法 最大熵算法 小波去噪 |
| DOI： |
| 分类号:TP391.41 |
| 基金项目:国家自然科学基金面上项目(61273215)； |
|
| IMAGE SEGMENTATION FOR WATERSHED ALGORITHM BASED ON WAVELET DENOSING AND GRADIENT CORRECTION |
|
Zhou Xiaoming
|
| Abstract: |
| Watershed segmentation method is an effective and commonly used segmentation method of image, but it is easy to cause the over segmentation and sensitive to noise.In order to overcome this problem,a kind of watershed image segmentation method based on wavelet transform and gradient correction is proposed in this paper.First, the wavelet is used to denoising,and then,using the maximum entropy algorithm to modify the gradient image.Finally the watershed algorithm is applied to the modified gradients to partition the image.Experimental results show that the proposed method achieves good image segmentation effect. |
| Key words: image segmentation watershed algorithm maximum entropy algorithm wavelet |
