Developing robust and cost-effective methods for benthic biodiversity assessments is crucial for mapping vulnerable habitats and designing effective management strategies. Achieving this objective requires increasing the spatial and temporal coverage of observations and optimizing the cost-effectiveness of sampling efforts. Remotely operated vehicles (ROVs) together with computer vision-based tools present great potential for cost-effectively mapping benthic taxa. However, the processing of hours of footage manually is time-consuming, making the process inefficient and costly. To address this challenge, an automated system for the detection and abundance estimation of 13 key benthic taxa was developed. The proposed approach analyses videos obtained with a ROV and provides taxon-specific counts to support habitat monitoring. It was developed and tested using a total of 42 sampling transects, resulting in approximately 20 hours of high-quality video footage. Two models were trained and compared for benthic taxa detection: YOLOv8x and Faster R-CNN. Their performance was evaluated using mAP50, reaching values of 0.782 ± 0.027 for YOLOv8x and 0.755 ± 0.032 for Faster R-CNN, indicating strong detection capability by both models. Since YOLOv8x showed a slightly higher performance with lower variance, it was selected and used for abundance estimation. Comparison of automated abundance estimates with manual counts across 10 transects gave a mean absolute percentage error (MAPE) of 20.4%, indicating a relatively high level of agreement between automated and expert-based estimations. Consequently, this approach enables non-invasive and cost-effective monitoring of benthic biodiversity through digital imagery while extending detection to taxa not previously considered, offering valuable insights for benthic habitat preservation.