Detection and imaging of subwavelength features in the subsurface using diffracted waves are rapidly gaining momentum in the oil and gas industry as well as in the fields of engineering, archeology, and homeland security. Most of the methods include coherent summation of the recorded wavefield along diffraction traveltime surfaces from point scatterers. The summation focuses energy onto point-like diffractors that appear at the resulting images as prominent anomalies. However, in cases in which the target is an elongated object such as a fault plane, fracture, tunnel, or elongated cave, a more efficient imaging method can be constructed. We have developed an algorithm for detecting and characterizing linear subsurface elements using a linear-diffractor operator. Our algorithm is based on the coherent summation of the edge diffraction generated by the entire lineament and on the analysis of the calculated coherence measure (semblance). The advantages and limitations of our method are discussed, and the results are compared to conventional point-diffractor-based techniques. Synthetic and real data examples demonstrate that using a linear-diffractor-based algorithm can dramatically improve the detection of linear objects.