In order to improve the speed of the robot in the daily life and improve the robot's cognitive environment, a scheme is proposed to detect unknown objects by means of the object surface roughness. This method launches a systematic method for determining discussion and analysis for local surface irregularities of the two type determining method for target objects, replaces the determination of Boolean to the continuous local surface irregularities, uses the two elements of that surface irregularities and law degree direction information to design a new class of segmentation weight calculation method, and then enter the scene to construct corresponding undirected weighted graph, after this, combines fast graph partitioning algorithm to obtain the corresponding target object. The experimental results show that: compared to the judgment method based on the irregularities convexity, the robustness of the method is better than that of the observation noise and the estimation error, the performance of the proposed method is sufficient to meet the needs of the practice based on the complexity of learning and reasoning methods, it is more in line with the needs of practice.
