Electrical Capacitance Tomography (ECT) image reconstruction algorithms utilize the sensitivity matrix as prior knowledge to invert the distribution of media within the pipeline, hence, the sensitivity matrix plays a crucial role in the quality of image reconstruction. Due to the soft-field characteristics of ECT sensors, the sensitivity distribution within the field is not uniform, with significant differences between the central and peripheral areas of the sensor, affecting the quality of the reconstructed images. This paper proposes a sensitivity preprocessing method based on the Sigmoid function. Simulation experimental results show that this method can significantly improve the uniformity of the sensitivity matrix and greatly enhance the quality of the reconstructed images: the Image Spatial Error (ISE) is reduced by 0.3825; the Relative Error (RE) is reduced by 0.386 5; and the Correlation Coefficient (CC) is increased by 0.0608, validating the feasibility of this method in ECT image reconstruction.