首页> 外文会议>2014 International Conference on Advances in Engineering and Technology Research >Content-aware dark image enhancement using fast fourier transform
【24h】

Content-aware dark image enhancement using fast fourier transform

机译:使用快速傅立叶变换的内容感知暗图像增强

获取原文
获取原文并翻译 | 示例

摘要

Image contrast enhancement is a typical problem in image processing and computer visualization. This problem increases for images taken under poor lighting conditions. So different enhancement technique is necessary to improve such images that are captured in extreme lighting conditions, such as very bright or dark environments. The enhancement task, however, is complicated by require of any general unifying theory of image enhancement. In this paper a content-aware algorithm is described that enhances dark images, sharpens edges and conserves the smoothness of flat regions. In this method the RGB colour image is firstly transformed to hue-saturation-value (HSV) colour space, then the algorithm is applied to enhance the image intensity (i.e. V component). To enhance the intensity value fast fourier transform is used on fixed size of cells (8×8) and then enhanced intensity values are normalized to overcome out of range problem.
机译:图像对比度增强是图像处理和计算机可视化中的典型问题。对于在不良照明条件下拍摄的图像,此问题会加剧。因此,需要不同的增强技术来改善在极端光照条件下(例如非常亮或黑暗的环境)中捕获的此类图像。然而,由于任何通用的图像增强统一理论的要求,增强任务变得复杂。在本文中,描述了一种内容感知算法,该算法可增强深色图像,锐化边缘并保留平坦区域的平滑度。在这种方法中,首先将RGB彩色图像转换为色相饱和度(HSV)色彩空间,然后将该算法应用于增强图像强度(即V分量)。为了增强强度值,对固定大小的像元(8×8)使用快速傅立叶变换,然后对增强的强度值进行归一化,以解决超出范围的问题。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号