基于改进CV模型的高原鼠兔图像分割

发布时间：2018-07-06 17:11

本文选题：高原鼠兔 + 图像分割　；参考：《华中科技大学学报(自然科学版)》2017年08期

【摘要】：针对CV模型无法分割灰度不均图像的问题,提出改进CV模型.在CV模型能量泛函中引入前景灰度不均抑制项,使CV模型能够对前景灰度不均图像进行分割.此外,在图像预处理阶段引入四叉树方法和大津法,缩小CV模型分割图像的搜索范围,减少分割时间;弥补CV模型须要手动设置初始轮廓的问题.实验结果表明:改进CV模型能够实现对前景灰度不均图像的精确分割,且耗时少,在背景抑制、目标区域轮廓定位等方面效果好.
[Abstract]:Aiming at the problem that CV model can not segment uneven grayscale image, an improved CV model is proposed. In the energy functional of CV model, the suppression term of foreground grayscale unevenness is introduced, so that the CV model can segment the foreground grayscale uneven image. In addition, the quadtree method and the Otsu method are introduced in the image preprocessing stage to reduce the searching range and the segmentation time of the CV model, and to make up for the problem that the CV model needs to set the initial contour manually. The experimental results show that the improved CV model can achieve accurate segmentation of foreground grayscale uneven images, and it takes less time, and has good effect in background suppression and target region contour location.
【作者单位】：兰州理工大学计算机与通信学院;
【基金】：国家自然科学基金资助项目(61362034,81360299)
【分类号】：TP391.41

【相似文献】