To develop intelligent robots, it is necessary to establish effective methods for visual and tactile recognition. Much of the existing work on tactile recognition has assumed that tactile sensors derive data only on object surfaces and generalizes the recognition algorithm. With tactile sensors which recognize local features of contact areas, e.g. vertices, edges etc., it is possible to improve interpretations more effectively than with existing methods. This paper proposes a method for object identification using local feature sensors. In the method, interpretations at each step of the identification process are represented as a set of subsets of 3-D space where a sensor is located. As the identification process progresses, the set becomes narrower. Consequently, local constraints become strict and inconsistent interpretations will be efficiently rejected. In the rejection, local constraints are compared considering the error bounds of the measurements in the sensor data.