My experience with BioRad image lab shows that normalisation work good only if the unstained blue ot mixture of unstained blue with Biorad all blue standarts is used as the software can recognize such standarts automatically, you will also need to use stain free compatible imager and gels. I never suceeded to produce anything usefull when doing things manually with i.e. Kaleidoscope protein standarts in non-stain free system. I also would like to note that normalisation don't affect the way the image looks it is only allow you to see more intricate detail in form of peaks that sometimes elusive when just bands are observed. Please refer to Biorad manual to perform normalisation for more details. Hope it helps with your project. 

Thank you very much for your help! I have the Kaleidoscope-marker and I already tried the normalization, however it is not that awesome... As we don't use stain free-equipment I will look for other ways, maybe via ImageJ.

use image j software it will give better band detection and band intensity as well as lane%