Abstract Channel networks exert a key control on drainage basins shape and dynamics, including the transfer of water and sediments throughout basins, and thus hydrosedimentary hazards. Landscape dissection by channels results from the competition between hillslope processes and channelized erosion processes such as overland flow or debris flows. In contrast to fluvial channelization, the transition from hillslopes to colluvial channels remains understudied, and high‐resolution LiDAR DEMs open new perspectives for the extensive extraction of channel heads. Several channel extraction methods exist but none is yet robust on fast eroding landscapes. Here we develop the CO 2 CHAIN method which identifies the hillslope to channel transition in drainage basins based on relative changes of local and upstream measures of flow convergence. We calibrate CO 2 CHAIN by fitting its results to channel head mapping made by geomorphologists in four contrasted basins in the United States and France with moderate to high erosion rates. Compared to two state‐of‐the‐art channel extraction methods, it fits best the experts' mapping in terms of average characteristics and variability without needing to be recalibrated and does not have to implement a drainage area threshold although it imposes a length scale threshold. This allows to revisit studies on channelization that have not yet included high erosion rate basins.