Accurate estimation of the State of Health (SOH) of lithium-ion batteries is extremely important for the reliability and safety of energy storage systems. This paper presents a way for estimating the SOH of lithium-ion batteries on the basis of Differential Thermal Voltammetry (DTV) and an integrated learning model fusion. Initially, the DTV curves are filtered, and health-related features are extracted and selected from the peak and valley positions. Subsequently, leveraging the learning speed and applicability of the Extreme Learning Machine (ELM), an integrated learning framework is proposed to minimize the prediction error of an individual model. At last, the dependability and robustness of the proposed way is verified by the Oxford University lithium-ion battery aging dataset.