With the development of Internet of Things (IoT) technology, the amount of sensitive data generated on surveillance camera increased drastically. However, the resources of surveillance camera for data privacy protection are limited. Therefore, how to protect surveillance video data privacy through limited resources become an important issue. In response to this problem, this paper proposes a hardware-oriented lightweight block cipher called ASEGF. The ASEGF algorithm has the Tpye-2 Extended Generalized Feistel Network (EGFN) structure, where different nonlinear components are used in the nonlinear layer of its round function, achieving a trade-off between security and resource consumption. Security analysis results show that the ASEGF algorithm can resist common cipher attacks. Avalanche effect test results show that the proposed algorithm has good diffusion and confusion properties. Performance analysis results show that, compared to the known lightweight algorithms, our proposed algorithm has lower resource consumption and faster implementation. This indicates that the ASEGF is suitable for use in surveillance video.