Recent progress in deep learning methods has shown that key steps in object detection and recognition can be performed with convolutional neural networks (CNN). In this article, we adapt YOLO (You Only Look Once) to a new approach to perform object detection on satellite imagery. This system uses a single convolutional neural network (CNN) to predict classes and bounding boxes. The network looks at the entire image at the time of the training and testing, which greatly enhances the differentiation of the background since the network encodes the essential information for each object. The high speed of this system combined with its ability to detect and classify multiple objects in the same image makes it a compelling argument for use with satellite imagery.
The latest advances in Deep Learning based methods and computational capabilities provide new opportunities for vehicle tracking. In this study, YO-LOv2 (You Only Look Once-version 2) is used as an open source Convolutional Neural Network (CNN), to process high-resolution satellite images, in order to generate the spatio-temporal GIS (Geographic Information System) tracks of moving vehicles. At first step, YOLOv2 is trained with a set of images of 1024 × 1024 resolution from the VEDAI database. The model showed satisfactory results, with an accuracy of 91%, and then at second step, is used to process aerial images extracted from aerial video. The output vehicle bounding boxes have been processed and fed into the GIS based LinkTheDots algorithm, allowing vehicles identification and spatio-temporal tracks generation in GIS format.
A reference database of addressing is an important resource for urban applications. The efficiency of an addressing system depends on both data quality and technical architecture. Data must respect a standard model that is flexible to meet different cases in the field. The technical architecture should be service oriented to offer a shared resource for multiple users and applications. This paper is to develop an addressing model for Morocco that extends Davis's and Fonseca's model presented in their work on the certainty of locations produced by an address geocoding system. We discuss the addressing data dictionary and acquisition plan in Morocco, revealing a diversified data management environment, characterized by multiple sources and actors. As a novelty in the field of GIS, we establish our technical architecture around cloud computing, according Service Oriented Application (SOA) standards. Our approach is based on the three pillars of cloud computing which are Software as a Service (SaaS), Platform as a Service (PaaS) and Infrastructure as a service (IaaS).
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.