The increasing spectrum of different cartilage repair strategies requires the introduction of adequate non-destructive methods to analyse their outcome in-vivo, i.e. arthroscopically. The validity of non-destructive quantitative ultrasound biomicroscopy (UBM) was investigated in knee joints of five miniature pigs. After 12 weeks, six 5-mm defects, treated with different cartilage repair approaches, provided tissues with different structural qualities. Healthy articular cartilage from each contralateral unoperated knee joint served as a control. The reflected and backscattered ultrasound signals were processed to estimate the integrated reflection coefficient (IRC) and apparent integrated backscatter (AIB) parameters. The cartilage repair tissues were additionally assessed biomechanically by cyclic indentation, histomorphologically and immunohistochemically. UBM allowed high-resolution visualisation of the structure of the joint surface and subchondral bone plate, as well as determination of the cartilage thickness and demonstrated distinct differences between healthy cartilage and the different repair cartilage tissues with significant higher IRC values and a steeper negative slope of the depth-dependent backscatter amplitude AIBslope for healthy cartilage. Multimodal analyses revealed associations between IRC and the indentation stiffness. Furthermore, AIBslope and AIB at the cartilage-bone boundary (AIBdC) were associated with the quality of the repair matrices and the subchondral bone plate, respectively. This ex-vivo pilot study confirms that UBM can provide detailed imaging of articular cartilage and the subchondral bone interface also in repaired cartilage defects, and furthermore, contributes in certain aspects to a basal functional characterization of various forms of cartilage repair tissues. UBM could be further established to be applied arthroscopically in-vivo.