In this paper multiple image super resolution is performed by using maximum likelihood estimation method in spatial domain. Various methods have been proposed to achieve multiple image super resolution, but most of the existing methods having drawbacks that they cannot reconstruct the edges present in an image properly and having inverse problem due to impulse and Gaussian noises. In this paper in order to preserve edges and to reduce Gaussian and impulse noises present in an image maximum likelihood estimation method is used in order to reconstruct the high resolution image effectively. The noises can be reduced by using Gaussian and linear filters and edges are preserved by using gradient descent method. The mean square error value of the input and output images are calculated. Finally the mean square error value of the output images are minimized by different iterations and high resolution image has been reconstructed with edge preservation

1. Changhyunkim, Kyuhachoi, kyuyounghwang, and Jongbeomra "Learning-based Super-resolution Using a Multi-resolution Wavelet Approach" School of Electrical Engineering and Computer Science, Kaist Year of 2012.
2. Chintan k. modi, Milan n. Bareja Electronics & Communication Engineering "An Effective Iterative Back Projection Based Single Image Super Resolution Approach" 2012 International Conference on communication systems and network technologies.
3. Guillaume Lemaitre, Heriot-watt university, Universitat de Girona, University De Bourgogne "Image analysis: an introduction to Super Resolution Using Wavelet. "
4. Gajjar, P. P. Dhirubhai Ambani Inst. of Inf. & Commun. Technol. , Gandhi agar, India Joshi, M. V. "New Learning Based Super-Resolution: Use of DWT and IGMRF Prior "Year of 2010.
5. H c liu1, Y Feng1 and G Y sun2. "Wavelet domain image super-resolution reconstruction based on image pyramid and cycle-spinning" Year of 2006.
6. Heng su, Liang tang, ying wu, senior member, IEEE Daniel Tretter, and Jie Zhou, Senior member, IEEE "Spatially Adaptive Block-Based Super-Resolution". IEEE Transactions On Image Processing, Vol. 21, No. 3, March 2012.
7. Min Chen1, Guoping Qiu1and Kin- Man Lam2. "Example Selective And Order Independent Super Resolution" Year Of 2012.
8. Shubinzhao, Huahan and Silong Peng "HMT-Based Image Super Resolution" Year Of 2003.
9. Weisheng Donga, b. Lei Zhangb, Guangming Shia, Xiaolin WUC "Image Deblurring And Super-Resolution By Adaptive Sparse Domain Selection and Adaptive Regularization" Year Of 2009.

Maximum Likelihood Estimation Gaussian Noises Impulse Noises Mean Square Error.