images: unnecessary images removed

notes/sections/1.md

@@ -20,7 +20,7 @@ an histogram and plotted with matplotlib. The result is shown in @fig:landau.
 
 ![Example of N points generated with the `gsl_ran_landau()`
 function and plotted in a 100-bins histogram ranging from -10 to
-80.](images/landau.pdf){#fig:landau}
+80.](images/landau-hist.png){#fig:landau}
 
 
 ## Randomness testing of the generated sample
@@ -86,7 +86,7 @@ Hence, the data was reasonably sampled from a Landau distribution.
 Contrary to what one would expect, the $\chi^2$ test on a histogram is not very
 useful in this case. For the test to be significant, the data has to be binned
 such that at least several points fall in each bin. However, it can be seen
-(@fig:Landau) that many bins are empty both in the right and left side of the
+(@fig:landau) that many bins are empty both in the right and left side of the
 distribution, so it would be necessary to fit only the region where the points
 cluster or use very large bins in the others, making the $\chi^2$ test
 unpractical.