Image subtraction with non-constant kernel solutions

It is demonstrated that non-constant kernel solution, that can fit the spatial variations of the kernel can be obtained with minimum computing time. The CPU cost required with this new extension of the image subtraction method is almost the same as for fitting constant kernel solution. The method is demonstrated with a serie of Monte-Carlo images. Differential PSF variations and differential rotation between the images are simulated. It is shown that this new method is able to achieve an optimal results even in these difficult cases. It is clear that the most frequents instrumental problems are automatically corrected by the method. It is also demonstrated that the method does not suffer problems with under-sampling of the images. To conclude the method is applied to real images of crowded field. It is proved that much larger sub-areas of the images can be used for the fit, while keeping the same accuracy in the subtracted image. This result is especially important in case of variables located in low density fields, like the Huchra lens. Many other useful applications of the method are possible for major astrophysical problems. For instance supernovas search should benefit the method, but other applications like Cepheids surveys in other galaxies should also find some advantage in using the method. Many other applications will certainly show-up, since variability search is a major issue in astronomy