过度分割足以令应用算法得到的结果变得毫无用处。
Application of over-segmentation algorithm is enough to make the results useless.
实验表明，改进后的算法在解决过度分割问题上，效果明显。
The experiment shows that the excessive segmentation problem can be solved using our improved algorithm efficiently.
但是，组合的聚类中心数目会多于实际的聚类数目，造成过度分割。
But the number of these combined clusters may be larger than that of the actual clusters, which may result in the over-segmentation.
针对分水岭算法产生的过度分割现象，结合一个模板进行了算法的改进。
Aiming at the over-segmentation phenomenon produced by using watershed algorithm, this paper improves this algorithm in virtue of a panel.
此外，为避免过度分割，还提出了一种方法用来合并过度分割出的场景。
In addition, we present a method to merge the over-segmented scenes.
分水岭算法用于图像分割时通常由于噪声的存在和梯度的不规则造成过度分割。
Due to noise and irregularity of the gradient image, watershed algorithm used to segment images generally leads to over-segmentation, which is unacceptable.
最后应用灰度齐次性准则和边界强度准则进行区域的融合，从而解决了过度分割问题。
Finally the application of gray-scale homogeneity criteria and guidelines for the border regions of intensity integration, in order to address the problem of over-segmentation.
为解决过度分割的问题，给出了区域的灰度特性和边界强度及区域尺寸相结合的融合准则。
To solve the problem of the over-segmentation, the criterions of region intensity and edge strength and region size are combined.
将该模型应用于图像分割，解决了传统模糊art网络图像分割结果随警戒参数的升高而出现的过度分割。
In the application of the image segmentation, the new model solves the over-segmentation of the original Fuzzy ART neural network algorithm due to the vigilance's increase.
在论文中使用了分水岭分割算法与区域合并相结合的方法，有效地对虹膜图像进行分割，并减少过度分割现象。
The method of combining watershed algorithm and re-gions merging technology is used in this paper, segment iris im-age accurately, and reduce over - segmentation phenomena efficiently.
针对传统分水岭变换对图像进行分割时常见的过度分割情况，提出了利用先验知识对梯度幅度图像进行处理的方法。
Using knowledge which has been known beforehand in Grads Range, this paper proposes an image segmentation method which can solve the problem.
针对传统分水岭变换对图像进行分割时常见的过度分割情况，提出了利用先验知识对梯度幅度图像进行处理的方法。
The images gradient magnitude mean and the variance of every pixel in the image relative to the gradient magnitude mean are calculated according to the method proposed in the paper.
文中首先利用快速均值漂移算法对图像进行过度分割，然后再利用改进的水平集方法进行分割， 最后由轮廓提取算法将感兴趣区域提取出来。
First, fast mean shift algorithm is used to over-segment the image. Second, improved LI-method is used to automatically segment the image.
它根据场景有效划分的特点，对分割结果进行评估，融合那些被过度划分的无效区域。
Based on the characteristics of a valid partitioning, it evaluates the partitioning result and merges the cells evaluated as invalid.
它根据场景有效划分的特点，对分割结果进行评估，融合那些被过度划分的无效区域。
Based on the characteristics of a valid partitioning, it evaluates the partitioning result and merges the cells evaluated as invalid.
应用推荐