Road Line Classification for Cartographic Generalization: a Neural Net Approach
نویسندگان
چکیده
This work studies the chance of using a neural network for classifying a set of road lines through a supervised learning process, trying to emulate a classification performed by a human expert. Here we present data segmentation, expert classification, data enrichment, and the neural network design, training and validation. The network selected is a feedforward backpropagation type with one hidden layer. To feed the input layer data enrichment consists of a set of characterizing quantitative measures derived from a principal component analysis. Qualitative information is also included in the form of administrative categories. Results quality is analyzed by means of error matrices after a crossvalidation process. The percentage of agreement is over 81%.
منابع مشابه
Kohonen Self Organizing for Automatic Identification of Cartographic Objects
Automatic identification and localization of cartographic objects in aerial and satellite images have gained increasing attention in recent years in digital photogrammetry and remote sensing. Although the automatic extraction of man made objects in essence is still an unresolved issue, the man made objects can be extracted from aerial photos and satellite images. Recently, the high-resolution s...
متن کاملLarge scale road network generalization for vario-scale map
The classical approach for road network generalization consists of producing multiple maps, for a different scale or purpose, from a single detailed data source and quite often roads are represented by line objects. Our target is the generalization of a road network for the whole scale range from large scale, where roads are represented as area objects, to mid and small scales, where roads are ...
متن کاملProperties Based Simplification of 2D Urban Area Map Object
For a long time, Geo-spatial information was printed on paper maps whose contents were produced once for specific purposes and scales. These maps are characterized by their portability, good graphic quality, high image resolution and good placement of their symbols and labels. These maps have been generated manually by cartographers whose work was hard and fastidious. Today, Computers are in us...
متن کاملB-spline Curve Fitting: Application to Cartographic Generalization of Maritime Lines
Generalization is the process of abstraction applied when the scale of a map is changed. It involves modifications of data in such a way that the data can be represented in a smaller space, while best preserving geometric as well as descriptive characteristics. A map is an abstracted model representing the geometric reality. The smaller the scale, the more schematic the representation. Line car...
متن کاملRoad Data Updating Using Tools of Matching and Map Generalization
It is one of the important ways for GIS data updating based on map generalization. This paper analyzes the main steps for road data updating based on map generalization. As the core of this updating process, matching method considering the levels analyses and selective omission based on mesh density are developed. The approach for road data updating based on these two tools is proposed, which i...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2005