| 段召蓬,彭文豪,李岚.基于图像分割和融合的图像去雾算法研究[J].井冈山大学自然版,2024,45(1):68-75 |
| 基于图像分割和融合的图像去雾算法研究 |
| RESEARCH ON IMAGE DEFOGGING ALGORITHM BASED ON IMAGE SEGMENTATION AND FUSION |
| 投稿时间：2023-07-10 修订日期：2023-11-03 |
| DOI：10.3969/j.issn.1674-8085.2024.01.001 |
| 中文关键词: 图像去雾 图像分割 图像融合 暗通道先验 自适应伽马校正 |
| 英文关键词: image defogging image segmentation image fusion dark channel prior adaptive gamma correction |
| 基金项目:陕西省自然科学基础计划项目（2021JM-399）；西安石油大学研究生创新基金项目(YCS22213187) |
|
| 摘要点击次数: 130 |
| 全文下载次数: 170 |
| 中文摘要: |
| 为了解决传统的暗通道先验去雾方法产生的细节丢失和亮度偏低等问题，本研究提出了一种基于图像分割和融合的去雾算法。首先对输入图像使用亮度反转的MSRCR预处理来进行色彩保真；其次用阈值分割法提取图像的特征信息并获得掩膜，根据特征信息设计自适应的Gamma校正方法，提升对比度和亮度，并使用暗通道先验方法保持去雾后的细节；最后将处理后的图像进行掩膜融合。在真实世界的数据集上仿真结果表明，本研究所提算法在去雾后能保留更多的细节且提高亮度。与几种经典的算法相比，本研究所提算法在去雾后的图像有较好的色彩保真度，保留了更多的细节，去雾效果好且亮度更自然。 |
| 英文摘要: |
| In order to solve the problems of details lost and low brightness caused by the traditional dark channel prior defogging method, a new defogging algorithm based on image segmentation and fusion is proposed in this paper. First, the input image is preprocessed by MSRCR of brightness inversion to achieve color fidelity, then the image features are extracted by threshold segmentation and the mask is obtained. An adaptive gamma correction method is designed to improve the contrast and brightness of the image. A method of dark channel prior is used to preserve the details after dehazing. The simulation results on real-world datasets are shown that the proposed algorithm in the paper can retain more details and improve the brightness after defogging. Compared with several classical algorithms, the proposed algorithm has better color fidelity, more details, better defogging effect and more natural brightness. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
