Computer vision is considered a useful tool for evaluating the external quality of fruits. Some solutions for capturing the entire surface image of fruits have been implemented, but they still have limitations, such as not being able to guarantee capturing the entire surface, bulky, or expensive. In this study, a two-shot and simple system for capturing the entire surface image of fruits was proposed. With the support of a mirror system, the top and lateral surfaces of the fruits were captured. To capture the bottom surface of the fruits, a flip mechanism has been integrated into the system. Testing results with pomelo and mango showed that the entire surface of the fruits was fully observed with two shots. This shows that the proposed system has great potential for imaging the entire surface of fruits. Besides, this solution can also be easily integrated into automatic inspection applications to evaluate the quality of agricultural and other products.