The invention relates to a capacitive tomography image reconstruction method based on a closed-loop control principle. The method mainly comprises the following steps of firstly, obtaining N(N-1)/2 independent measurement values under an empty field condition and N(N-1)/2 independent measurement values under the condition that an object exists through measurement by utilizing a capacitive tomography imaging system, and establishing an N*N capacitive variation matrix by utilizing a difference between the two types of the independent measurement values; secondly, taking a Calderon algorithm as acontrolled object, solving a new capacitive variation matrix by utilizing dielectric constant distribution inversely derived by the algorithm to serve as a negative feedback, and correcting a deviation value of capacitive variation by adjusting parameters of a PID controller, thereby enabling a whole closed-loop to be converged; and finally, if the number of iterations reaches a preset number ofiterations, ending the iteration and outputting a reconstruction result, otherwise, continuing the iteration. According to the method, the closed-loop control principle is combined with the Calderon algorithm for the first time, so that an image reconstruction error can be effectively reduced and the image reconstruction quality is remarkably improved; and the method has important practical valuesand good application prospects in the field of electrical imaging.