The phenomenon of bioclogging in porous media is widely present in nature and engineering and is closely related to fields such as the environment, energy, and biomedical applications. In this work, we focus on the bioclogging process in porous media and investigate the effects of flow rate and pore size on the behavior of biofilm clusters and the evolution of permeability using a microfluidic chip-microscope-CMOS camera visualization experimental system. By integrating the microparticle image velocimetry technique, we achieved real-time dynamic observation of the flow field within porous media. Flow-visualization experimental results show that flow rate and pore size control the surface morphology and ultimate clogging efficiency of biofilms by influencing shear rates and nutrient exchange rates. It is shown that bioclogging in porous media presents two distinct clogging patterns, characterized by pattern Ⅰ with preferential flow paths and pattern Ⅱ without preferential flow paths. Pattern Ⅰ occurs under conditions of a smaller flow rate and larger pore size, where the fluid mainly flows concentrately in the preferential flow paths. The unevenness of the flow rate distribution is exacerbated over time, affecting the stability of clogging, and the permeability decline shows obvious intermittent fluctuations. Pattern Ⅱ occurs under conditions of a higher flow rate and smaller pore size, and the flow field distribution is relatively uniform with no obvious high-speed concentrated areas. In this pattern, the clogging effect is much more significant, with the permeability ratio being reduced by three orders of magnitude at the end of the experiments.