| 宽动态红外图像增强算法综述 |
| |
| 引用本文： | 周永康,朱尤攀,曾邦泽,胡健钏,欧阳慧明,李泽民.宽动态红外图像增强算法综述[J].激光技术,2018,42(5):718-726. |
| |
| 作者姓名： | 周永康 朱尤攀 曾邦泽 胡健钏 欧阳慧明 李泽民 |
| |
| 作者单位： | 1.昆明物理研究所, 昆明 650223 |
| |
| 摘 要： | 红外成像系统设计中，一般采用宽动态的采集电路以获得丰富的细节信息，当前大部分的显示设备都只有8位，所以将宽动态图像压缩成低动态图像同时保持尽可能多的信息成为重点。研究了当前主流的宽动态红外图像处理算法，分析了映射、图像分层和梯度域3类算法的优缺点，实现了3类算法中的经典算法，并用同一张红外图片进行对比分析，提出了各类算法的改进意见。图像分层算法要在抑制光晕和梯度反转的情况下降低时间复杂度；梯度域算法需在进一步提高细节信息的情况下抑制背景噪声。
|
| 关 键 词： | 图像处理 红外图像增强 宽动态算法 映射 图像分层 梯度域 |
| 收稿时间： | 2017-11-01 |
Review of high dynamic range infrared image enhancement algorithms |
| |
| Abstract: | In order to obtain rich image detail information, a wide dynamic acquisition circuit is often used in an infrared imaging system. However, most of the display devices are only 8 bits, how to compress wide dynamic images into low dynamic images and preserve detail information as much as possible is one of the key technologies of infrared image display. Current mainstream infrared image processing algorithms, such as mapping, image stratification and gradient domain are introduced, and the advantages and disadvantages of the 3 algorithms are analyzed. Contrast analysis is made with the same infrared image, and the improved opinions of all the algorithms are proposed. Image stratification algorithm needs to reduce time complexity in the case of inhibition of halo and gradient inversion. Gradient domain algorithm needs to suppress background noise and then the details are further improved. |
| |
| Keywords: | |
|
| 点击此处可从《激光技术》浏览原始摘要信息 |
|
点击此处可从《激光技术》下载全文 |
|
