José Luis García Balboa

Generalization-oriented Road Line Classification by Means of an Artificial Neural Network

In line generalization, a first goal to achieve is the classification of features previous to the selection of processes and parameters. A feed forward backpropagation artificial neural network (ANN) is designed for classifying a set of road lines through a supervised learning process, attempting to emulate a classification performed by a human expert for cartographic generalization purposes. The main steps of the process are presented in this paper: (a) experimental data selection; (b) segmentation of lines into homogeneous sections, (c) sections enrichment through a set of quantitative measures derived from a principal component analysis, and qualitative information derived from road network and road type; (d) expert classification of the sections; and finally (e) the ANN design, training and validation. The quality of results is analyzed by means of error matrices after a cross-validation process giving a goodness, or percentage of agreement, over 83%.In line generalization, a first goal to achieve is the classification of features previous to the selection of processes and parameters. A feed forward backpropagation artificial neural network (ANN) is designed for classifying a set of road lines through a supervised learning process, attempting to emulate a classification performed by a human expert for cartographic generalization purposes. The main steps of the process are presented in this paper: (a) experimental data selection; (b) segmentation of lines into homogeneous sections, (c) sections enrichment through a set of quantitative measures derived from a principal component analysis, and qualitative information derived from road network and road type; (d) expert classification of the sections; and finally (e) the ANN design, training and validation. The quality of results is analyzed by means of error matrices after a cross-validation process giving a goodness, or percentage of agreement, over 83%.

Generalization-oriented road line segmentation by means of an artificial neural network applied over a moving window

In line generalization, results depend very much on the characteristics of the line. For this reason it would be useful to obtain an automatic segmentation and enrichment of lines in order to apply to each section the best algorithm and the appropriate parameter. In this paper we present a methodology for applying a line-classifying backpropagation artificial neural network (BANN) for a line segmentation task. The procedure is based on the use of a moving window along the line to detect changes in the sinuosity and directionality of the line. A summary of the BANN design is presented, and a test is performed over a set of roads from a 1:25k scale map with a recommendation of the value of the parameters of the moving window. Segmentation results were assessed by an independent group of experts; a summary of the evaluation procedure is shown.

Sinuosity pattern recognition of road features for segmentation purposes in cartographic generalization

In line generalization, results depend very much on the characteristics of the line in question. For this reason it would be useful to obtain an automatic segmentation and enrichment of lines in order to apply to each section the best algorithm and the most appropriate parameter. In this paper, we present a line segmentation methodology based on a sinuosity pattern recognition measured by means of the effective-area as derived from the Visvalingam-Whyatt algorithm. Sections are determined by applying the Douglas-Peucker algorithm to a shape signature of the line: an effective-area/length space representation. An experiment is carried out with a set of 24 road features from a 1:25000 scale map with a recommendation of the value of some parameters and a procedure for the automated search of that defined as natural number of sections. This procedure is based in the search of zones of stability in a graph of the number of sections when applying Douglas-Peucker to the shape signature. The results are positively assessed by an independent group of experts.

Analysis of User and Producer Risk when Applying the ASPRS Standards for Large Scale Maps

Using a statistical simulation process, the behavior of the Accuracy Standards for Large Scale Maps (ASLSM) of the American Society for Photogrammetry and Remote Sensing is analyzed according to the sample size, as well as the relation between the limiting errors (thresholds), stated by the standard as a root mean squared error, and the actual root mean squared error of the product. When the root mean squared error of the product equals the threshold of the standard the simulation results show the ASLSM is very restrictive, classifying 75 percent of products as Class 2 maps instead of Class 1 maps. If the variability of the product is greater or lesser than this threshold, results can be depicted by a family of acceptance curves. These curves can be employed by users to determine the sample size needed to limit their acceptance risk, but also by producers to analyze their rejection risk.

Approximating Cartography to the Customer's Expectations: Applying the “House of Quality” to Map Design

The design of a map and guide for a Spanish natural park has been guided by the application of a product-development methodology known as quality function deployment (QFD). QFD is a tool for bringing the voice of the customer into the product-development process, from conceptual design to manufacturing. In order to develop a high-quality product whose design meets customers’ needs, market research has been developed to discover customers’ expectations and the strengths and weaknesses of competitors’ products. Sixteen main customer expectations (WHATs) were considered in relation to product comfort, content, and portrayal. In order to take into account the aforementioned expectations, 24 technical descriptors (HOWs) were considered. The product was finally specified by all the technical descriptors and their target values (HOW MUCHs). Results of the methodology are expressed using a set of matrices that depicts a house, the “House of Quality,” that concentrates the most important aspects of a product plan....