An approach for rapid, automatic detection of plant roots in minirhizotron images is presented. The problem is modeled as a Gibbs point process with a modified Candy model, in which the energy functional is minimized using a greedy algorithm whose parameters are determined in a data-driven manner. The speed of the algorithm is due in part to the selection of seed points, which discards more than 90% of the data from consideration in the first step. Root segments are formed by grouping seed points into piecewise linear structures, which are further combined and validated using geometric techniques. After root centerlines are found, root regions are detected using a recursively bottom-up region growing method. Experimental results from a collection of diverse root images demonstrate improved accuracy and faster performance compared with previous approaches.