Multiplicative decomposition based image contrast enhancement method using PCNN factoring model

Contrast enhancement helps human perceive and analyze low level quality images and has an important role in image processing applications. Being different from traditional methods based on histogram processing, this paper explored an image contrast enhancement technique based on multiplicative decomposition. With improved pulse coupled neural network factoring model (PCNN-FM) specially designed for contrast enhancement application, images with different details were output as multiplicative factors. Initial input image could be rebuilt almost without deviation by multiplying these factors. So enhancing factor would improve corresponding image details. Based on this point, the paper divided initial image with the first factor output by proposed PCNN-FM to implement contrast enhancement. This processing equals rebuilding initial image with factor images by maximizing the first factor. Experiments show that the dynamic range of enhanced image is moderate, and detail information is rich. In addition, the proposed method is universal, easy to implement and has actual using value.

[1]  Jason M. Kinser,et al.  Theoretical foundation of the intersecting cortical model and its use for change detection of aircraft, cars, and nuclear explosion tests , 2004, Signal Process..

[2]  Reinhard Eckhorn,et al.  Feature Linking via Synchronization among Distributed Assemblies: Simulations of Results from Cat Visual Cortex , 1990, Neural Computation.

[3]  Liu Ji-fang.  Application of low-pass filtering and gray adjusting to image enhancement , 2012 .

[4]  M. L. Padgett,et al.  Multiscale image factorization , 1997, Proceedings of International Conference on Neural Networks (ICNN'97).

[5]  John L. Johnson,et al.  PCNN models and applications , 1999, IEEE Trans. Neural Networks.

[6]  Yvonne Herz.  Image Processing Using Pulse Coupled Neural Networks , 2016 .

[7]  Arthur R. Weeks,et al.  Practical considerations on color image enhancement using homomorphic filtering , 1997, J. Electronic Imaging.

[8]  M. L. Padgett,et al.  Pulse coupled neural networks (PCNN) and wavelets: biosensor applications , 1997, Proceedings of International Conference on Neural Networks (ICNN'97).

[9]  M.R. Alsharif,et al.  Image Enhancement Based on Logarithmic Transform Coefficient and Adaptive Histogram Equalization , 2007, 2007 International Conference on Convergence Information Technology (ICCIT 2007).

[10]  Zia-ur Rahman,et al.  Retinex processing for automatic image enhancement , 2004, J. Electronic Imaging.

[11]  Hao Ying,et al.  New algorithms for contrast enhancement in grayscale images based on the variational definition of histogram equalization , 2008, Integr. Comput. Aided Eng..

[12]  Yi-de Ma,et al.  An image segmentation based method for iris feature extraction , 2008 .

[13]  Luan Mengjie.  Image Enhancement of Finger-vein Patterns based on Median Filtering , 2012 .

[14]  Jason M. Kinser,et al.  Image Processing using Pulse-Coupled Neural Networks , 1998, Perspectives in Neural Computing.

[15]  J. L. Johnson.  Pulse-coupled neural nets: translation, rotation, scale, distortion, and intensity signal invariance for images. , 1994, Applied optics.

[16]  Zhang Rui.  Noise Removal in Color Images on Image Gradient , 2010 .