Bark beetles have become an escalating concern in Mexico’s forests, as extreme and highly variable climate conditions have turned these insects into a significant infestation threat. Currently, rangers manually count beetles after they are trapped, capturing images with their cellphones and sending them to specialized teams. These teams count and classify the beetles to point establishment strategies under the Mexican Official Norm NOM-019-SEMARNAT-2006. Nevertheless, the delay between image acquisition and expert analysis can extend to over a couple of weeks, enough time for beetle populations to grow into an infestation if the actions are not taken in time. To address this delay, we proposed a segmentation and classification pipeline designed to provide a preliminary assessment of the captured images, flagging those with concerning beetle counts. Segmentation was achieved using Connected Component Labeling, while classification was performed through a Convolutional Neural Network, both showing a favorable performance.