An Algorithm for Describing User Behavior Model of Remote Sensing based on User Profile Technology

Le résumé fourni par la source

At present, with the rapid development of satellite remote sensing platform, the quality of acquired satellite images has been greatly improved, and the amount of image data has also increased geometrically. For users, how to get useful image information quickly and accurately from massive remote sensing image data is the primary issue to improve the application efficiency of satellite data. The existing remote sensing data sharing platform has already done "what to do", but there is still a certain gap from what the user needs "what to give", and cannot truly realize the service on demand. Based on this, this paper adopts the concept of user portrait (User Profile) in large data analysis, and designs a remote sensing user information description method, which uses "user identification + time + behavior type + contact point" as the allocation weight model of information elements, describes user tags, and then uses the "foreground" in image processing field. The concept of "background" uses the LDA model to accurately define the user's portrait content. In the experiment, we used the actual data sharing platform user behavior information, detailedly enumerated the quantitative data of user portrait, and gave the follow-up application direction. Experiments show that the feasibility of the algorithm has certain application prospects.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.