多门限图像重叠目标区域精细分割方法仿真
|
摘要
研究有效的目标区域精细分割方法,能够降低分割消耗时间、提高分割目标区域内部的均匀性和目标区域间对比度,增强目标区域分割质量,在实际应用中具备一定使用价值。为了减少图像出现模糊、重影及分割误差较大等问题,提出基于Otsu法的多门限图像重叠目标精细分割方法,该方法是通过利用高斯函数平滑灰度直方图内的白噪声,利用平滑的灰度直方图来计算灰度平均值,获取多门限值;运用多门限值计算目标区域类之间总方差的离散度测度,获得最优有限门值,对该门限值进行赋值,将图像重叠目标区域像素值与该门限值进行对比,得到最优的分割重叠目标区域结果。实现多门限图像重叠目标区域精细分割。仿真结果证明,所提方法可以增强目标区域分割对比度,提高目标区域内部均匀性,减小分割距离误差率,提高消耗时间,增强图像分割质量,具一定的应用价值。
In order to reduce the image blurring, double image and large segmentation error, a method of fine segmentation for overlap target in multi-threshold image based on Otsu method was put forward. This method used Gaussian function to smooth the white noise in gray histogram, and then used smooth gray histogram to calculate the average gray value and obtain multi-threshold value. After that, this method used the multi-threshold value to calculate the dispersion measure of total variance between target region classes, so as to get the optimal finite threshold value. Moreover, we assigned this threshold value and compared the pixel value of overlap target region with this threshold value. Finally, we obtained the optimal result of segmenting split overlap target region. Thus, we realized the fine segmentation of multi-threshold image overlap target region. Simulation results prove that the proposed method can enhance the segmentation contrast of target region and improve the internal uniformity of target region. Meanwhile, this method can reduce the error rate of segmentation distance and enhance the image segmentation quality, which has a certain application value.
In order to reduce the image blurring, double image and large segmentation error, a method of fine segmentation for overlap target in multi-threshold image based on Otsu method was put forward. This method used Gaussian function to smooth the white noise in gray histogram, and then used smooth gray histogram to calculate the average gray value and obtain multi-threshold value. After that, this method used the multi-threshold value to calculate the dispersion measure of total variance between target region classes, so as to get the optimal finite threshold value. Moreover, we assigned this threshold value and compared the pixel value of overlap target region with this threshold value. Finally, we obtained the optimal result of segmenting split overlap target region. Thus, we realized the fine segmentation of multi-threshold image overlap target region. Simulation results prove that the proposed method can enhance the segmentation contrast of target region and improve the internal uniformity of target region. Meanwhile, this method can reduce the error rate of segmentation distance and enhance the image segmentation quality, which has a certain application value.
引文
[1] 陈晓春,张彦彬,彭虎,等. 重叠纤类图像的凹点匹配和分割算法[J]. 纺织学报, 2017,38(11):143-149.
[2] 王蕊超,邵景峰,白晓波,等.基于RSSI测距改进的多区域自适应室内定位方法[J].西安工程大学学报, 2017,31(3):402-410.
[3] 张永德,彭景春,刘罡,等. 基于水平集的前列腺磁共振图像分割方法研究[J]. 仪器仪表学报, 2017,38(2):416-424.
[4] 李为华,苏辉,郭华平. 融合多特征和谱聚类集成的图像分割方法[J]. 信阳师范学院学报:自然科学版, 2017,30(4):638-641.
[5] 袁恒东. 基于标签传递图割的图像分割算法[J]. 计算机技术与发展, 2017,27(12): 35-38.
[6] 刘松涛,刘振兴,姜康辉. 基于模糊Renyi熵和区域增长的图像目标分割方法[J]. 系统工程与电子技术, 2018,40(8):1693-1701.
[7] 彭晏飞,周娟,訾玲玲. 于卫星地域遥感图像分割方法研究仿真[J]. 计算机仿真, 2017,34(1):332-335.
[8] 李磊,董卓莉,张德贤,等. 基于区域限制的EM和图割的非监督彩色图像分割方法[J]. 电子学报, 2016,44(6):1349-1354.
[9] 叶青,胡昌标. 一种改进的基于图论的图像分割方法[J]. 计算机与现代化, 2016,10(9): 64-67.
[10] 史小丹,马国锐,陈王丽,等. 基于模糊阈值和水平集的红外图像分割方法[J]. 激光与红外, 2016,46(1):109-114.
[11] 林庆达,陈敏,禤亮,等. 图像分割方法应用于施工现场物体的识别[J]. 云南电力技术, 2017,45(3):63-67.
[12] 王志忠,王永增,马连成,等. 基于分水岭和形态学重构的矿石图像分割方法[J]. 物联网技术, 2017,7(3):89-91.
[2] 王蕊超,邵景峰,白晓波,等.基于RSSI测距改进的多区域自适应室内定位方法[J].西安工程大学学报, 2017,31(3):402-410.
[3] 张永德,彭景春,刘罡,等. 基于水平集的前列腺磁共振图像分割方法研究[J]. 仪器仪表学报, 2017,38(2):416-424.
[4] 李为华,苏辉,郭华平. 融合多特征和谱聚类集成的图像分割方法[J]. 信阳师范学院学报:自然科学版, 2017,30(4):638-641.
[5] 袁恒东. 基于标签传递图割的图像分割算法[J]. 计算机技术与发展, 2017,27(12): 35-38.
[6] 刘松涛,刘振兴,姜康辉. 基于模糊Renyi熵和区域增长的图像目标分割方法[J]. 系统工程与电子技术, 2018,40(8):1693-1701.
[7] 彭晏飞,周娟,訾玲玲. 于卫星地域遥感图像分割方法研究仿真[J]. 计算机仿真, 2017,34(1):332-335.
[8] 李磊,董卓莉,张德贤,等. 基于区域限制的EM和图割的非监督彩色图像分割方法[J]. 电子学报, 2016,44(6):1349-1354.
[9] 叶青,胡昌标. 一种改进的基于图论的图像分割方法[J]. 计算机与现代化, 2016,10(9): 64-67.
[10] 史小丹,马国锐,陈王丽,等. 基于模糊阈值和水平集的红外图像分割方法[J]. 激光与红外, 2016,46(1):109-114.
[11] 林庆达,陈敏,禤亮,等. 图像分割方法应用于施工现场物体的识别[J]. 云南电力技术, 2017,45(3):63-67.
[12] 王志忠,王永增,马连成,等. 基于分水岭和形态学重构的矿石图像分割方法[J]. 物联网技术, 2017,7(3):89-91.