摘要
提出了一种新的基于SOFM融合的亮度不均图像分割方法。首先利用基于SOFM网络求取最佳曝光值,然后在此曝光值附近选取两个曝光值获取两幅图片;再对各图片进行基于双向加权序贯平滑的二值化;最后将二值化后的图像进行基于SOFM的融合,这样就完成了亮度不均图像的分割。此方法分割能够很好地保持目标图像原有形态,噪声小,为亮度不均图像的分割提出了一种新的思路。
A method of ununiformity brightness image segmentation based on SOFM network fusion is proposed. First- ly, the best exposure value is got and two exposure values near the best exposure value are chosen to get two images using SOFM network. Then, the binarization method based on two-way weighting Sequential smooth is used to segment the two images which are obtained in different light intensity. Lastly, the two binary images based on SOFM fusion. After that, the final segmentation image is obtained. Using this method can record the information of objects accurately. The experimental results show that this method is much better and provides a way to segment the ununiformity brightness images.
出处
《舰船电子工程》
2015年第12期143-146,共4页
Ship Electronic Engineering
关键词
SOFM
融合
亮度不均
图像分割
SOFM, fusion, ununiformity brightness, image segmentation
