计算机工程与应用 ›› 2012, Vol. 48 ›› Issue (1): 5-7.

• 博士论坛 • 上一篇    下一篇

PCNN模型在彩色图像增强中的应用

韩丽娜1,2,耿国华1,周明全3   

  1. 1.西北大学 可视化技术研究所,西安 710127
    2.咸阳师范学院 图像处理研究所,陕西 咸阳 712000
    3.北京师范大学 信息科学与技术学院,北京 100875

  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2012-01-01 发布日期:2012-01-01

PCNN’s application in real color image enhancement

HAN Lina1,2, GENG Guohua1, ZHOU Mingquan3   

  1. 1.Institute of Visualization Technology, Northwest University, Xi’an 710127, China
    2.Institute of Graphics and Image Processing, Xianyang Normal University, Xianyang, Shaanxi 712000, China
    3.School of Information Science and Technology, Beijing Normal University, Beijing 100875,
    China
  • Received:1900-01-01 Revised:1900-01-01 Online:2012-01-01 Published:2012-01-01

摘要:

应用灰度图像增强方法对真彩图像进行增强,往往都会产生色彩偏离,影响增强结果和视觉效果。因此基于人眼视觉系统对亮度变化比较敏感,提出在HSV色彩空间,应用PCNN模型进行增强的方法。将真彩图像由RGB空间变换到HSV空间,保持色度H和饱和度S不变,结合入射反射模型,利用脉冲耦合神经网络(PCNN),对亮度V通道进行增强处理。将HSV空间得到的增强图像变换到RGB空间。实验证实,对一些对比度低、细节不明显的图像应用此方法进行增强,色彩基本无偏差,细节部分明晰,动态范围压缩较好,视觉效果得到了较大的改善。

关键词: 脉冲耦合神经网络, 真彩图像, 图像增强

Abstract:

Using gray image enhancement methods to enhance real color image can produce color deviation and bad visual effect. Therefore this paper presents the color image enhancement method which is based on PCNN model with characteristics of the human visual system more sensitive to the change of brightness in HSV space. It transforms the image from RGB space to HSV space and keeps the H and S channel unchanged. Pulse coupled neural network is used to process the V channel with illumination-reflectance model. Transforming the image to RGB space is executed. The experiment confirms that it can get a better result with color basic deviation, clear detail and better dynamic range using this method to enhance some images with low contrast, unclear details. The visual effect is greatly improved.

Key words: pulse coupled neural networks, real color image, image enhancement