“…To date, such compilations are somewhat cumbersome and time‐consuming, and must be dealt with very carefully. Although semi‐automated processes facilitating feature extraction are already in use (Aubry, Maturana, Efros, Russell, & Sivic, 2014; Chen, Chen, & Tseng, 2016; Chen et al, 2018; Fanos, Pradhan, Mansor, Yusoff, & bin Abdullah, 2018; Garosi, Sheklabadi, Conoscenti, Pourghasemi, & Van Oost, 2019; Goshen & Shimshoni, 2008; Hartley & Zisserman, 2003), they primarily refer to maps and photographs, leaving aside illustrations and paintings. More importantly, they devote less attention to synthesizing various features originating from different types of visual sources on a unified GIS‐based framework.…”