The invasive alien plants that cause ecological, economic, and public health disturbances are highly reproductive and thus must be controlled at the early stage of development. This study is intended to establish the system to construct the invasive alien plant distribution map based on spatial DBMS (Database Management System), obtain ortho-images using drones, classify invasive alien plants using CNN (Convolution Neural Network)-based image learning, and present how to generate the invasive alien plant distribution map using them. We obtained ortho-images by photographing the area where Sicyos angulatus, one of the invasive alien plants, inhabited in two test areas of Daejeon to examine the applicability of this system. In the first experiment, we conducted the image learning and classification using only the images photographed in test area a. In the second experiment, we conducted the image learning using the image photographed in test area a and classified the images photographed in test area b. In the last experiment, we conducted learning using the images photographed in test area a and the imagse of Sicyos angulatus on web sites and classified the image photographed in test area b. The analysis results showed the average classification accuracy of 95%, 45%, and 61% in the first experiment, the second experiment, and the third experiment, respectively. Construction of the distribution map of invasive alien plants requires the metatable of the ortho-images that contain various invasive alien plants. Additional information on the search targets of invasive alien plants in consideration of the expansion of new invasive alien plants can help to manage them with the spatial DBMS and carry out the automated classification of various invasive alien plants and to manage various ecosystem disturbances. Using them in the analysis system would help control invasive alien plants.