Using geometric constraints to improve performance of image classifiers for automatic segmentation of traffic signs

Automatic detection and recognition of traffic signs from images is an important topic in many applications. At first, we segmented the images using a classification algorithm to delineate the areas where the signs are more likely to be found. In this regard, shadows, objects having similar colours, and extreme illumination changes can significantly affect the segmentation results. We propose a new shape-based algorithm to improve the accuracy of the segmentation. The algorithm works by incorporating the sign geometry to filter out the wrong pixels from the classification results. We performed several tests to compare the performance of our algorithm against those obtained by popular techniques such as Support Vector Machine (SVM), K-Means, and K-Nearest Neighbours. In these tests, to overcome the unwanted illumination effects, the images are transformed into colour spaces Hue, Saturation, and Intensity, YUV, normalized red green blue, and Gaussian. Among the traditional techniques used in this study, the best results were obtained with SVM applied to the images transformed into the Gaussian colour space. The comparison results also suggested that by adding the geometric constraints proposed in this study, the quality of sign image segmentation is improved by 10%–25%. We also comparted the SVM classifier enhanced by incorporating the geometry of signs with a U-Shaped deep learning algorithm. Results suggested the performance of both techniques is very close. Perhaps the deep learning results could be improved if a more comprehensive data set is provided.

La susceptibilité aux glissements de terrain dans la ville d’Al Hoceima et sa périphérie : application la méthode de la théorie de l’évidence

Résumé : Le Rif Marocain en général et la ville d’Al Hoceima et sa périphérie urbaine plus particulièrement, connaissent fréquemment des aléas géomorphologiques, notamment les glissements de terrain qui entravent la gestion urbaine. Ce type d’aléa naturel est de grande actualité, aussi bien sur le plan scientifique que sur le plan médiatique, à cause de l’augmentation de la vulnérabilité, due aux circonstances des changements globaux (réchauffements climatiques) et à la forte urbanisation, souvent irrationnelle. L’objectif de cet article est la mise en place d’une approche objective visant l’évaluation de la susceptibilité aux glissements de terrain dans la ville d’Al Hoceima et sa périphérie. La théorie de l’évidence, qui est une méthode probabiliste bivariée, est fondée sur les règles de Bayes qui consistent à calculer la probabilité d’occurrence spatiale de glissements de terrain, en se basant sur la notion de probabilité à priori et de probabilité à posteriori, tout en considérant les glissements de terrain comme variable à modéliser et chaque facteur causatif comme variable prédictive. Le but de ce travail est de procéder à un zonage d’aléa glissement de terrain tout en assurant une bonne prédiction de ce phénomène avec une bonne résolution spatiale. Les résultats de la courbe de ROC (Receiver Operating Characteristic) montre que la confrontation de la carte de susceptibilité, des glissements de terrain à la carte d’inventaire, permet une capacité de prédiction considérable (AUC=0,889). Ceci pousse au constat selon lequel, plus de 2/3 des glissements de terrain inventoriés s’inscrivent dans des classes de susceptibilité élevée et très élevée. Ce produit cartographique peut constituer un puissant outil d’aide permettant la formulation des suggestions, dans le but d’optimiser l’évaluation du risque de glissements de terrain dans les zones exposées à ce phénomène. Mots clés : SIG, Théorie de l’évidence, Susceptibilité aux glissements de terrain, Al Hoceima (Maroc)

Quantifying Urban Expansion Using Landsat Images and Landscape Metrics: A Case Study of the Halton Region, Ontario

Urbanization is considered as one of the main factors affecting global change. The Halton Region as part of the Great Toronto Area (GTA), is regarded as one of the fastest growing regions in Canada, generating 20% of national GDP. It is also one of the most desirable places for living and thriving business. This research attempts to assess the urban expansion in the Halton Region, Ontario, Canada from 1989 to 2019 using satellite images, analysis approaches and landscape metrics. Multi-temporal Landsat images, and the supervised learning algorithms in GIS software were used to explore the dynamic changes, and to classify the urban and non-urban areas. The temporal urban expansion in the Halton Region experienced a dramatic rise, and mainly occurred from the centre of the area. The analysis of landscape metrics based on different methods, including Land Use in Central Indiana (LUCI) model, Vegetation-Impervious Surface-soil (V-I-S) model, and the census data of Canada was carried out to understand the transition mode of the urbanization in the Halton Region. Also, the population growth in the centre of the Halton Region was considered as one of driven forces affecting urban expansion. The results showed that most of the landscape metrics rose between 1989 and 2019, indicating leapfrog pattern of urbanization occurred over the entire period. The contribution of this research is to evaluate the urbanization in the Halton Region, and give the city managers a clear mind to make appropriate decisions in further urban planning.

SAR Phase Unwrapping Using Region-Growing with Polynomial-Based Phase Prediction

Phase Unwrapping for Synthetic Aperture RADAR Interferometry (InSAR) remains a challenge due to the speckle noise and temporal decorrelation present in many interferograms. This paper proposes a Polynomial-Based Region-Growing Phase Unwrapping (PBRGPU) approach that builds from the Region-Growing Phase Unwrapping (RGPU) approach developed by Xu and Cumming in 1996 (Xu and Cumming, 1996). This approach iteratively performs phase unwrapping at the edges of multiple seeded regions using a least-squares polynomial phase prediction, and conducts statistically rigorous quality assurance to identify low quality pixels from further processing. The approach uses a desired statistical confidence interval as its main parameter, which is more intuitive to users than other threshold parameters. The proposed approach is currently the only phase unwrapping approach to take this strategy with its quality assurance. The proposed approach improved upon the solution quality of the RGPU approach, in some cases achieving a tenfold decrease in RMSE for simulated data. Applying the proposed approach to RADARSAT-2 data collected over Polar Bear Provincial Park in Northern Ontario, Canada yielded positive results, and the PBRGPU approach consistently performed on par with or outperformed SNAPHU in terms of solution quality. The PBRGPU approach does lag behind SNAPHU in terms of the domain of the solution, with SNAPHU unwrapping a significantly larger portion of the interferogram in all test cases, but this issue could be mitigated through post-processing the unwrapped interferogram. The proposed approach provides a solid foundation for adaptive region-growing algorithms that integrate all available information rather than relying on pre-processing strategies.

Typologie d’un geoweb des flux et réseaux / Typology of a flow and network geoweb

RESUMÉ. L’analyse de la dynamique des aires urbaines ou des métropoles, la délimitation de leurs aires fonctionnelles et la comparaison spatio-temporelle de leurs motifs est souvent freinée par l’insuffisance de données relationnelles (portant sur des liens entre des entités) ouvertes et l’absence jusque récemment de dispositifs d’analyse et de géo-visualisation dédiés. Au-delà des questions d’ouverture des données (géo)numériques, nous proposons un panorama du geoweb, le processus de création de cartes dans le contexte du web 2.0, spécifique aux flux et réseaux localisés. L’éclairage ainsi apporté sur les pratiques cartographiques actuelles révèle trois grandes familles d’applications web ainsi que les besoins d’une communauté, restreinte mais dynamique, d’analyser librement ses propres jeux de données. ABSTRACT. Analysing the dynamics of urban areas or metropolises, delineating their functional areas and comparing their spatio-temporal patterns is often limited by the lack of open relational data (on links between entities) and the absence until recently of dedicated analysis and geo-visualization frameworks. Beyond the questions of opening (geo)digital data, we propose a panorama of a geoweb, the process of creating maps in the context of the Web 2.0, specific to flows and networks. The insights provided on current mapping practices reveal three main families of web applications, as well as the needs of a small but dynamic community to freely analyze its own data sets.

Geo-feature modeling uncertainties in discrete global grids: a case study of downtown Calgary, Canada

The Open Geospatial Consortium has officially adopted discrete global grid systems (DGGS) as a new option for Earth reference standards. Many state-of-the-art DGGS implementations have been developed, revealing the potential for DGGS applications. Before the wide application of DGGS in solving real-world problems, however, the potential uncertainties of modeling on DGGS should be investigated and documented. This study focused on the uncertainties of geo-feature modeling on DGGS, quantitatively measured the point position displacement and line and polygon features’ geometry distortion, and evaluated the validity of topological relationships. Specifically, traffic cameras (points), main streets (lines), and land-cover classes (polygons) of downtown Calgary (AB, Canada) were modeled in various DGGS configurations at multiple resolutions. Results showed that the point displacement and polygon distortion generally reduced when being modeled at a higher resolution. The tessellations with the monotonical convergence characteristic are recommended if cell indices are expected to represent levels of model precision. Line features’ fidelity was affected by grid tessellations, resolution levels, grid orientation relative to the Earth, and the rotated line directions. The degree of the line distortion was not straightforward to forecast. Maintaining the topological validity between spatial objects with various granularities was challenging and needed further algorithm development for DGGS implementations. The study outcomes can serve as useful guidelines in the selection among grid types, refinement ratios, and resolution levels when applying DGGS implementations to urban environments. This paper also pinpoints several research directions that can benefit the quantization and analysis of vector features on DGGS.

Estimer des flux de navetteurs avec un modèle gravitaire : application géomatique en région Provence-Alpes-Côte d’Azur (France)1

Cet article réinvestit la modélisation des flux de déplacement domicile–travail au moyen des modèles gravitaires de flux. À partir d’une demande émanant de la région Provence-Alpes-Côte d’Azur, un modèle répondant au principe de parcimonie, basé sur des données publiques ouvertes et facilement mobilisables, a été établi afin de répondre au besoin de prévision tout en assurant la transparence et la réplicabilité de la démarche. À partir de ce modèle, différentes hypothèses (type de variable sur les masses des entités géographiques, variations des calculs des courtes distances à vol d’oiseau, contrainte de portée spatiale due au budget-temps de déplacement, ré-échantillonnage aléatoire) ont été modélisées dans une optique de prospective et d’aide à la décision. Les résultats montrent qu’il est possible d’obtenir un modèle assez robuste avec les données disponibles de l’Institut national de la statistique et des études économiques (INSEE) et un modèle gravitaire log-linéaire, tout en réduisant sensiblement la taille de l’échantillon traité.

Ontologie d’Alerte Choucas : de la modélisation des connaissances à un outil support d’un raisonnement géovisuel

Lorsqu’une intervention de secours est nécessaire, localiser précisément et rapidement le site sur lequel envoyer les équipes est primordial. La littérature montre que des outils de géovisualisation constituent des solutions pertinentes pour supporter des processus d’analyse d’informations dans des contextes variés. Nous nous intéressons ici au raisonnement du secouriste réceptionnant un appel à l’aide et visons des solutions conceptuelles et logicielles dédiées à la tâche de détermination de la localisation de la victime, plus particulièrement dans le contexte du secours en montagne. Nous avons formalisé le raisonnement du secouriste et les informations sur lesquelles il ou elle s’appuie à l’aide d’une ontologie. L’Ontologie d’Alerte Choucas structure les concepts exploités par le secouriste qui élabore des hypothèses de localisation probable de la victime à partir d’informations (telles qu’une position relative, un temps de marche, une direction) fournies lors d’un échange verbal. Dans notre approche, l’ontologie est en outre exploitée pour dériver les composants d’interface d’un prototype de géovisualisation facilitant le raisonnement du secouriste. Ces composants sont une aide à la saisie des informations, en fournissant une restitution cartographique adaptée, et contribuent à construire et à affiner la zone de localisation. Notre approche présente une chaine de traitements originale, menant de la représentation des connaissances à la génération automatisée d’une interface fonctionnelle d’aide au raisonnement visant à localiser des victimes en montagne.

Geovisualization and harmonic analysis for the exploratory search of localized cyclic recurrences in spatio-temporal event data

Many geovisualization environments integrate graphical representations of time. Some of them include representation of both linear and cyclic aspects of time, providing an exploratory analysis of spatio-temporal data through several temporal cyclic scales. However, few of them provide an exploratory analysis of localized cyclic recurrences in spatio-temporal data. Ad hoc temporal diagrams, representing both linear and cyclic aspects of time, provide a visual search for cyclic recurrences in temporal data when the possibility is left to the user to perform a gradual modification of the represented cyclic scale’s duration. The combination of these graphic representations of time, with cartographic representations, displaying the spatial distribution of such cyclic recurrences, could provide an exploratory analysis of localized cyclic recurrences in spatio-temporal data. Mathematical tools coming from other scientific fields, such as the harmonic analysis, offer another way to identify cyclic behaviors in temporal data. Combining the visual approach offered by specifically designed geovisualization environments, with a harmonic analysis that suggests searching paths to the user during its exploratory analysis, can then improve the visual search for localized cyclic recurrences. We propose a geovisualization environment, which combines, on one hand, a visual analysis of localized cyclic recurrences in spatio-temporal data, using ad hoc temporal diagrams, cartographic representations, and specific semiologic rules, and on the other hand, mathematical tools, such as harmonic analysis and spatial clustering, that provide searching paths to the user for its visual analysis. This approach is supported by a geovisualization environment, GrAPHiST, which provides an exploratory analysis of spatio-temporal event data.