This paper presents a tactile-sensing fingertip based on piezoelectric materials for edge shape detection. Two statistical features were extracted from the collected tactile signals and evaluated through ML algorithms such as the Support Vector Machine (SVM) and One-Dimensional Convolutional Neural Network (1D-CNN). The system was tested with two 3D-printed cubes featuring Bar and Roof edges, which were rotated in the hand to assess the tactile sensing system capability in edge detection. Results showed that the SVM algorithm attained an accuracy of 90.21%, with 87.93% for CNN and 87% for human perception. The proposed system demonstrated its effectiveness in discriminating between object edges, highlighting its potential for applications in robotics and prosthetics.