One of the structures of greatest interest in the maxillofacial region is the mandibular canal, where its preservation optimizes the success of clinical procedures. Due to its disposition inside the mandible, in vivo visualization requires the use of radiographic techniques, among which the Cone Beam CT (CBCT) stands out. The objective of this work is to compare the visualization of the human mandibular canal by means of CBCT images with different observers and radiation programs. Using images generated with CBCT in dry mandibles with different acquisition programs, coronal slices were obtained in the anterior, middle and posterior sectors of the mandibular canal, with a total of 240 measurements, which included two voxel sizes and two doses different radiation sources. Each of the images was evaluated by a Dentist, an Anatomy Master's degree and a specialist in Oral & Maxillofacial Radiology. Percentages of visualization according to observer were obtained for each image acquisition program. The Cochran Q test yielded differences between the observers (p <0.05) and the McNemar test defined these dissimilarities between the specialist and the other observers, but not between the latter. The Yates and Mann-Whitney U tests did not show statistical differences according to the acquisition program (p> 0.05). The degree of expertise of the observer influences the visualization and its precision. In turn, using an image acquisition program with a different voxel/radiation dose not affect the visualization, since it depends on the anatomy of the structure itself. A protocol with less radiation could be used in patients complying with the ALARA (“As Low As Reasonably Achievable”) premise.