An image processing apparatus capable of communicating with a plurality of servers stores image data including an object of recognition, and a plurality of recognition dictionaries. The image processing apparatus establishes communication with one of the servers to receive, from the server with which the communication has been established, designation information designating a recognition dictionary for recognizing the object of recognition included in the image data. The image processing apparatus identifies the recognition dictionary designated in the received designation information from among the stored recognition dictionaries and uses the identified recognition dictionary to recognize the object of recognition included in the image data.