To improve the target location accuracy of a small-scale airborne electro-optical platform, a target geo-location algorithm based on extended Kalman filtering (EKF) is proposed. According to the characteristics of a tracking target locked by the airborne electro-optical platform, the same target is measured repeatedly. Using the aircraft position and attitude information measured by an integrated navigation system as well as the position information of the gimbal angles from the position encoder, the direction of the line of sight for the target is determined according to the Earth ellipsoid model. The state and measurement equations are established, and the geographical position of the target is estimated using EKF. The Monte Carlo method is used to analyze the influence of the measurement error on the target geo-location accuracy. The simulation results demonstrate that the proposed algorithm is highly accurate and robust. The validity of the algorithm is verified by a flight test. At a flight height of 4300 m, the geo-location error of the target is less than 15 m. Compared with that of the algorithm based on the Earth ellipsoid model, the target geo-location accuracy of the proposed algorithm is improved obviously.