For through-the-wall-radar signal processing, there exist extensive studies on removing the wall surface reflection signal, while how to eliminate/alleviate the in-wall structure reflection is not well addressed. In many building structures, a layer of reinforced steel bars and utility pipes exist inside the wall which can cause strong clutter to overwhelmingly mask the reflection signal from the targets under test behind the wall. Such clutter cannot be mitigated using the conventional wall clutter removal methods. Thus, a new effective technique to remove the strong inside-wall rebar or pipe reflection is indispensable. Considering the correlated features of the in-wall rebar or pipes and the spatial sparsity of the behind-wall targets under test, a low-rank and sparse representation model-based in-wall clutter suppression algorithm is developed in this letter for target feature enhancement and detection. Experiments on both simulation data and field test data are performed for performance evaluation and validation.