Abstract: In real-world，incomplete spatial data often have a heavy effect on the spatial decision，analysis and reasoning.So far，many statistical methods have been developed to deal with such incomplete items.However，these methods are not able to detect the incomplete spatial data since they ignore the spatial characteristics，e.g.spatial dependency among spatial data.This paper proposes a k-Nearest-Neithbors（k-NN） based approach for detection of incomplete spatial data，where the 3σ edit rule is used to identify the incomplete spatial data in the k-NN of each of spatial samples.Finally，a practical example is used to illustrate the rationality of the proposed approach in this paper.