The construction sector produces more than one-third of the world's solid waste. Construction and demolition waste (CDWs) are generated from the construction, renovation and demolition of buildings, roads, bridges and other structures. Moreover, CDW include the materials that may suddenly be generated by natural disasters, such as earthquakes and floods. Post-earthquake building waste (PBW) is typically composed of a mixture of different materials, such as concrete, bricks, tiles, ceramics, wood, glass, gypsum and plastic. These materials represent, if properly separated, a high potential for recycling and reuse particularly the inert fraction, representing about 70% of the total. From this perspective, this work aims to develop an innovative strategy based on optical sensing in order to identify and classify different types of PBW coming from a post-earthquake site (Amatrice, Italy). A strategy based on hyperspectral imaging (HSI) working in the SWIR range (1000-2500 nm) was developed. The acquired hyperspectral images were analyzed using different chemometric methods: principal component analysis (PCA) for data exploration and partial least-squarediscriminant analysis (PLSDA) to build a classification model. Results showed that the proposed approach allows to recognize and classify inert fractions from contaminants (i.e., wood, plastics and drywall). The obtained results show how HSI could be particularly suitable to perform classification in complex scenarios as produced by earthquakes.