| 基于MSR和融合理论的X射线安检图像增强算法 |
| |
| 引用本文： | 郑林涛,董永生,史恒亮.基于MSR和融合理论的X射线安检图像增强算法[J].电视技术,2015,39(7). |
| |
| 作者姓名： | 郑林涛 董永生 史恒亮 |
| |
| 作者单位： | 河南科技大学信息工程学院,河南洛阳,471023 |
| |
| 基金项目： | 国家自然科学基金,基于剪切波域核子空间建模的纹理表示与检索研究(61301230)；河南省教育厅科学技术研究重点项目，基于方向多尺度变换和有限混合模型的纹理识别方法（13B520992） |
| |
| 摘 要： | 针对X射线安检图像噪声大、对比度低和边缘不清晰等问题,提出一种结合多尺度Retinex算法(Multi-Scale Ret-inex,MSR)算法和图像灰度最大值融合的双重能量X射线图像增强算法.首先,应用MSR算法对高能和低能X射线图像进行处理得到初步增强结果,然后采用空域灰度值最大融合算法融合经过MSR算法初步增强后的高能和低能X射线图像从而得到最终增强X射线图像.实验结果表明该算法能更有效地提高双重能量X射线图像的对比度,显著改善图像质量.
|
| 关 键 词： | X射线图像 图像增强 多尺度Retinex算法 图像融合 |
| 收稿时间： | 2014/4/23 0:00:00 |
| 修稿时间： | 2014/5/17 0:00:00 |
An x-ray security inspection image enhancement algorithm based onMSR and fusion theory |
| |
| ZHENG Lintao,DONG Yongsheng and SHI Hengliang.An x-ray security inspection image enhancement algorithm based onMSR and fusion theory[J].Tv Engineering,2015,39(7). |
| |
| Authors: | ZHENG Lintao DONG Yongsheng and SHI Hengliang |
| |
| Affiliation: | Information Engineering College,Henan University of Science and Technology,Information Engineering College,Henan University of Science and Technology,Information Engineering College,Henan University of Science and Technology |
| |
| Abstract: | Considering the characteristic of X-ray security inspection images such as much noise, low image contrast and blurred edges, a dual energy X-ray image enhancement algorithm is proposed. This algorithm is based on MSR(Multi-Scale Retinex) and image grayscale the maximum value integration. First of all, the high energy and low energy X-ray images is transformed with MSR as the initial enhanced result. Then the initial enhanced result is fused with spatial gray level value method and the fused result is the final enhanced image. Experimental results show that the algorithm can more effectively improve dual-energy X-ray image contrast and significantly improve image quality. |
| |
| Keywords: | X-ray image image enhancement Multi-Scale Retinex image fusion |
| 本文献已被 万方数据 等数据库收录！ |
| 点击此处可从《电视技术》浏览原始摘要信息 |
|
点击此处可从《电视技术》下载全文 |
|
