Object-based land cover mapping and comprehensive feature calculation for an automated derivation of urban structure types at block level

Abstract

Las investigaciones sobre la evolución morfológica de las ciudades colombianas se concentran en ciudades fundadas durante la época colonial española. El objetivo de la presente investigación fue caracterizar las formas urbanas y su relación con los espacios públicos, a partir de la topografía, la transformación ambiental, los procesos de gestión urbana y los desarrollos urbanísticos entre 1849 y 2017. Mediante el uso de un sistema de información geográfica (SIG), se hizo el análisis de superposición de capas, así como el mapa de pendientes y la caracterización de laderas. Se seleccionaron 24 barrios de los 114 que tiene en total Manizales. Como resultado se identificaron 8 tipologías de tramas urbanas, diferenciadas en 5 fases de crecimiento urbano. La trama de damero y los grandes movimientos de tierra se dejaron de emplear desde mediados del siglo XX. Las tramas posteriores tratan de adaptarse a las laderas con menos modificación del terreno, pero sin reducir el riesgo geotécnico. En cuanto al espacio público, muchas de las zonas verdes son laderas con espacios residuales que no cuentan con diseño de espacio público y no permiten el uso recreativo, al contrario de áreas localizadas en topografías planas, que favorecen el uso recreativo y sin tratamientos geotécnicos. Palabras clave: desarrollo urbano; ecología urbana; espacio público; gestión del riesgo; morfología urbana; ordenamiento territorial; sistema de información geográfica (SIG);

Urban Land Use Mapping by Combining Remote Sensing Imagery and Mobile Phone Positioning Data

“…High-spatial-resolution RSI has shown promise as a data source for obtaining urban land use maps, because it can provide both spectral and textural properties of urban areas [10][11][12]. Various spectral-spatial classification methods have been proposed to extract urban land use maps.…”

Contributing to WUDAPT: A Local Climate Zone Classification of Two Cities in Ukraine

“…The world urban database and access portal 60 tools (WUDAPT) initiative (http://www.wudapt.org) is work-61 ing toward the goal of mapping the LCZs of all major cities 62 globally [4], [5]. 63 There is a considerable literature emerging on the use of 64 remote sensing to classify cities according to urban structure 65 types (USTs) [6]- [8], also referred to as urban morphology 66 types [9] and urban structural units [10]. However, as pointed 67 out in [6], most of the previous studies have analyzed only 68 one city with little thought for transferability to other areas.…”

“…63 There is a considerable literature emerging on the use of 64 remote sensing to classify cities according to urban structure 65 types (USTs) [6]- [8], also referred to as urban morphology 66 types [9] and urban structural units [10]. However, as pointed 67 out in [6], most of the previous studies have analyzed only 68 one city with little thought for transferability to other areas. 69 Each has their own classification scheme, which renders mul-70 ticity comparisons impossible.…”

Multiscale Geoscene Segmentation for Extracting Urban Functional Zones from VHR Satellite Images

“…Existing UFZ analysis with VHR satellite images is always regarded as a computer-vision task, and focuses mainly on developing scene classification methods [52,53]. In these cases, UFZs are usually represented by image tiles or road blocks [22,23,26]. However, UFZs can have different shapes and sizes, and should be analyzed at different scales [54]; thus, existing studies are weak in UFZ representations.…”

“…They utilized roads as boundaries to segment VHR satellite images into blocks which were regarded as functional zones. However, this method will be ineffective in the following three cases: (1) when the segmented blocks contain different kinds of functional zones [25], so a block does not fall entirely within an individual functional zone; (2) when there are temporal differences between VHR images and the used road data [26]; and (3) when researchers are seeking functional zones at a scale other than that of the blocks. Functional zones usually have different sizes and heterogeneities, which should be derived from segmentation results at multiple scales [27].…”