A novel method based on curved space field is applied to coal pile recognition is proposed

in order to avoid coal pile accident in coal mines.Firstly

the internal concave information is detected with luminance signal and entropy distribution feature in coal pile image.The gradient feature in concave and convex area is described by curved quantum.The gray information is removed from image.Then all the sub-curved space

which is formed of curved quantum group

constituted the curved space field by linking up with each edge of curved quantum smoothly.Finally

the coal pile model is structured by degree of polymerization.The early warning based on the coal pile height and radius is ready for the upper limit

which is calculated according to the coal pile model.This method can overcome the effects of noise and blurred image on the recognition of coal pile with taking advantage of cured space field with the resistance to environmental noise.Experiments show that the proposed application has a better detection speed with 26.3 frame per second and the accuracy is 98.2 %.