| 1. | Recently single - threshold or multi - threshold is often used to segment image and detect object contour on an image by means of genetic algorithm
本文首先对传统遗传算法进行了改进，提出了适应函数标定公式，定义了相似度概念。 |
| 2. | In final , segmented images were analyzed , and the areas of all different sections of the image were calculated ( presented with pixel number )
论文最后对分割后的图象进行分析，计算出了图象中各部分所占的面积(用象素数表示) 。 |
| 3. | We propose a low - dimensional region - based shape index to retrieve images . the initial step in our approach is to segment images into regions on dominant colors
首先基于主要颜色把图象分割成区域，然后把分割后得到的区域用作形状检索的输入。 |
| 4. | This algorithm carries out pretreatment of ordinary military graphs , then segments images , extract edges in the margin , and finally conducts feature matching
该算法实现了对普通军事图像的预处理,然后进行图像分割、提取边缘,最后进行了特征匹配。 |
| 5. | Firstly , human skin - color information is used to segment images into skin color region and non - skin color region , then pretreat the skin - color region and come into the phase of image similarity
首先利用彩色图像的颜色信息把彩色图像分割成肤色区域与非肤色区域，然后对肤色区域做预处理后，进入图像似然度检测阶段。 |
| 6. | A method based on differential operators is presented for treating sophisticated contours . the algorithm of extracting directional contours from segmented images is also discussed . semi - automatic scheme is proposed
我们提出了一种新方法，它利用边界微分算子来尽量在自动分类过程中保留图像特征，还讨论了轮廓线的定向提取技术。 |
| 7. | Gmrf is a kind of segmentation methods which combine gray characteristic and texture characteristic as its characteristic . gmrf segments image effect . in this method , to combine the mrf and the characteristic which has the guass distribution is the key , and we discussed in the paper
Gmrf分割作为一种结合灰度特征与纹理特征的分割方法在图象分割中得到很好的效果，在gmrf分割方法中，讨论了mrf与guass分布结合条件。 |
| 8. | The images to be processed is obtained in condition of nonuniform illumination . a threshold surface by interpolating zero - crossing points is used to segment image to avoid influence of nonuniform illumination . then the destination region can be detected using the method of pattern recognition and the method advoid the influence of nonuniform illumination effectively . the contour description is obtained by fitting at least squares principle
论文在分析当前图象分割的现状和趋势的基础上，采用基于log算子的动态阈值选取对图象进行分割，以克服非均匀光场给图象分割带来的影响，然后通过对区域特征的提取和分类准则的确定，准确识别出目标区域。 |
| 9. | A developed simple m - s model for image segmentation in geometric active contour model is presented based on intra - region similar and inter - region dissimilar properties . the model constructs an energy ( cost ) function , which is made of intra - region variations and weighting squares of subtraction of region mean values . using gradient - descent methods , the energy function is minimized and we get a curve evolution equation that segments image
基于区域内一致性加权区域间差异性构造能量函数,利用最陡梯度法使能量函数最小化,提出了一种改进的简化mumford - shah ( m - s )图像分割模型,该模型利用区域内方差描述区域内一致性,区域间平均灰度值之差的平方描述区域间差异性,实验结果表明,通过调节加权系数,该模型对弱边界图像分割具有较强的适应性。 |
| 10. | The developed simple m - s model is suitable for weak edge image segmentation by adjusting weighting coefficient . the experiments on tumor ct image segmentation show that geometric active contour model requires more computational load than snake model but it can converiently control the topology construct of segmented image
在肿瘤ct医学图像分割实验中,通过比较参数活动轮廓模型与几何活动轮廓模型可知,参数活动轮廓模型计算量小于几何活动轮廓模型计算量,而几何活动轮廓模能适应拓扑结构的变化。 |
