No full-text available
To read the full-text of this research,
you can request a copy directly from the author.
Correlations between census dwelling data and remotely sensed data
Abstract
With the increasingly available census data and remotely sensed data, to discuss their relationship is one of important issues in GIS data integration. This paper proposes a method to demonstrate the correlations between zonal census dwelling data and residential densities discriminated by RS classification. First, texture statistic (homogeneity) along with six TM bands (bands 1-5 and 7) is put together to classify residential density levels and it is shown that the homogeneity enhances classification accuracy. After classification, close correlations between residential densities and census dwelling data have been examined by using multiple linear regression. However, the accurate classification scenario does not methodically reveal higher correlations. It is concluded that data integration of zonal census data within the framework of RS-GIS is feasible, consequently the differentiation of residential densities could offer enormous opportunities to treat dwelling-related census data, such as identification of the scale and zonal effect of the modifiable area unit problem (MAUP), and topographical representation of zonal census data.
... Concerning methods such as spatial disaggregation of population, the key is to combine such methods with earth observation data and remote sensing techniques in order to achieve fully integrated urban system models (Bracken and Martin, 1989;Sim, 2005;Steinnocher et al., 2006). With the increasingly available census data and remotely sensed data, to discuss their relationship is one of important issue in GIS data integration (Chen, 1998;Chen, 2002). Aubrecht et al. (2009) have demonstrated how disaggregated population data can improve estimation of exposure to earthquake hazard. ...
In order to discuss in detail the environment and urban systems, it is necessary to consider not only static physical objects like buildings, but also the spatiotemporal aspects like the distribution of population. This paper aims to construct models that describe the spatiotemporal distribution of population in urban areas. The models are composed of parameters describing the number of persons per unit floor-Area of buildings, which varies according to the time fluctuation factors and the location factors, and are calibrated using a person trip survey data and GIS data. We discuss the characteristics of the spatiotemporal distribution of population and the accuracy of the models, and demonstrate that the proposed models can benefit all phases of urban planning, which include risk assessment and disaster management.
... Texture is produced by an aggregation of unit features that may be too small to be discerned separately on the image (Lillesand & Kiefer, 2000). Texture classification and texture algorithms are different from traditional statistical methods because they rely on statistical properties of a neighbourhood of pixels (Chen, 1998). For example, local (neighbourhood) intensity variance or other statistics derived from individual pixel attributes are sometimes used as statistical texture measures (Musick & Grover, 1992). ...
... For urban studies and especially for hazard and risk analyses the inclusion of population in these models is essential. For example, Chen [131] describes correlations between census dwelling data and remotely sensed data, and Banzhaf et al. [132] detect negative growth in the city of Leipzig by integrating remote sensing data. ...
Cities are complex systems composed of numerous interacting components that evolve over multiple spatio-temporal scales. Consequently, no single data source is sufficient to satisfy the information needs required to map, monitor, model, and ultimately understand and manage our interaction within such urban systems. Remote sensing technology provides a key data source for mapping such environments, but is not sufficient for fully understanding them. In this article we provide a condensed urban perspective of critical geospatial technologies and techniques: (i) Remote Sensing; (ii) Geographic Information Systems; (iii) object-based image analysis; and (iv) sensor webs, and recommend a holistic integration of these technologies within the language of open geospatial consortium (OGC) standards in-order to more fully understand urban systems. We then discuss the potential of this integration and conclude that this extends the monitoring and mapping options beyond "hard infrastructure" by addressing "humans as sensors", mobility and human-environment interactions, and future improvements to quality of life and of social infrastructures.
... photographs. The next step in urban system modeling is the integration of earth observation data with ancillary spatial and space-related information. This enables the transition of land cover to specific land use ( Mesev, 2005). For urban studies and especially for hazard and risk analyses the inclusion of population in these models is essential. Chen (1998) speaks about correlations between census dwelling data and remotely sensed data. Banzhaf, Kindler, and Haase (2006) detect negative growth in the city of Leipzig by integrating remote sensing data and statistical data. Anselin (1999) and Weeks (2001) explain the role of spatial analysis in social sciences and demographic research and sp ...
Integrative analysis of remote sensing data and socioeconomic information enables the transition of land cover and urban structures into a detailed functional model of urban land use. In this paper object based image analysis is used to derive a classification of urban structures. The implementation of ALS (Airborne Laser Scanning) significantly enhances the classification of optical imagery both in terms of accuracy as well as automation. Land cover types are additionally differentiated based on their relative height above ground resulting in a 3D building model. This model forms the basis for the integration of socioeconomic data for identifying urban functions. Buildings are split into sub-buildings by creating Thiessen polygons based on geocoded address point data. Company data is linked to this address information resulting in significant refinement of the functional classification and concrete identification of building use. By means of spatial disaggregation, raster population data is distributed to potential residential buildings. The relevant potential residential capacity is calculated under consideration of building use and ALS-based height information. These additional information sources guarantee a high accuracy of disaggregation and a further refinement of the functional 3D city model, independently confirmed by a quantitative accuracy assessment.
... Texture is produced by an aggregation of unit features that may be too small to be discerned separately on the image (Lillesand & Kiefer, 2000). Texture classification and texture algorithms are different from traditional statistical methods because they rely on statistical properties of a neighbourhood of pixels (Chen, 1998). For example, local (neighbourhood) intensity variance or other statistics derived from individual pixel attributes are sometimes used as statistical texture measures (Musick & Grover, 1992). ...
Aim Ecological factors that vary along spatial scales can greatly influence the outcome of evolution. However, often it is difficult for researchers to identify significant ecological variables that might be associated with the geographical distribution of phenotypes and genotypes in nature. In this paper, we use remote sensing image texture analyses to investigate breeding territories of the white‐throated sparrow ( Zonotrichia albicollis ), a polymorphic passerine species that has been shown to segregate spatially according to morph. Our aim was to examine how an integrative measure of landscape feature, such as image texture, could be used to distinguish subtle differences in habitat use and, presumably, habitat selection.
Location A population of white‐throated sparrows located 1.5 miles east of the Cranberry Lake Biological Station in the Adirondack Mountains of St Lawrence County, New York, was used in this study; the study site itself was approximately 1 km ² .
Methods We analysed 27 territories (16 white males; 11 tan males) to determine if image texture can be used to distinguish differences in vegetation and structure in territories of the two colour morphs. First (variance)‐ and second (homogeneity)‐order image texture measurements with three different window sizes were derived using the green, red, and near infrared bands of Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) data. Texture was then placed into t ‐tests to determine statistical differences between the territories of the two male morphs.
Results Significant differences between the territories of the two morphs were found in both variance and homogeneity texture measures in the red and green ASTER bands in all three window sizes. There were no differences with any near infrared texture measures.
Main conclusions Examination of first‐ and second‐order statistical measures indicated that the territories of tan males have significantly more spectral variance than the territories of white males, while white males’ territories are significantly more homogeneous. These findings are consistent with the previous finding of the morphs settling in ‘high’ and ‘low’ neighbour density habitats that differ with respect to the amount of useable area on their borders. Previous study has also suggested that white and tan male territories differ according to overall landscape structure, not vegetation composition. Therefore, we suggest that red and green spectra may be more useful when examining differences in territory structure and possibly territory quality of similar edge‐type species. Our results suggest that image texture is a useful tool for ecological and evolutionary modelling of the distributions of phenotypes/genotypes across landscapes.
In recent years, demographics regarding the spatiotemporal distribution of people in urban areas estimated from location information of mobile phone users are available for various usage. However, the available information of mobile phone demographic data is limited due to various reasons.
In order to address this limitation, we firstly constructed a model to estimate the spatiotemporal distribution of people by considering the differences of population density according to the location, time, and uses of building where they are. Next, we complemented their attribute and purpose of stay for each grid cell, by incorporating the information of Person Trip survey data (PT data). Finally, using the advanced spatiotemporal demographic data, we discussed the spatiotemporal distribution of people, which varies according to areas, date, and time. Summary and conclusions are as follows.
(1) First, we proposed a model to estimate the number of people being inside/outside of buildings for each building use in detailed units of space and time, by using Mobile Spatial Statistics (MSS) data and Building Point data, which is the GIS data including detailed attribute information such as floor area and building use.
(2) Next, we added the detailed attribute information to the people being inside/outside of buildings for each building use, by using the information of Person Trip survey data (PT data) in which detailed personal attributes as well as the location and time information of the departure and arrival, purpose of trip, and means of transportation are available. In this process, we utilized the fact that detailed personal attributes of people are deeply dependent on the location, time, and building use where they are staying.
(3) Finally, using the spatiotemporal distribution of people which was estimated by the proposed model, we attempted to making regional comparisons of the spatiotemporal distribution of people in urban areas.
In conclusion, we demonstrated that it is possible to grasp the spatiotemporal characteristics of population distribution attached with detailed attribute information such as their age, gender, occupation, and purpose of stay that vary according to the location, time, and building use where they are.
Recently, it is possible to grasp the spatiotemporal distribution of people in cities using population statistics based on the location information of mobile phone users. However, it is difficult to know their purpose of stay which varies according to the use of building they stay and their detailed attributes such as age and gender. In this paper, we firstly propose a model that describes the number of people staying inside/outside of buildings by considering the population density that varies according to the use of building, time, and local characteristics, by using GIS database and Mobile Spatial Statistics (MSS) which is one of the population statistics of mobile phone users. Next, we integrate the MSS data and the Person Trip survey data (PT data) which include detailed personal attributes as well as the purpose of stay. Using the integrated database, we demonstrate the advanced use of population statistics based on mobile phone users by addition of purpose of stay which varies according to building use.
Satellite images are a convenient and effective tool to measure landscape patterns because they provide a digital mosaic of the spatial arrangement of land covers (Chuvieco, 1999). However, satellite remote sensing data have not been applied to the urban landscape to the same degree that they have to other landscapes because the ability to discern urban features from satellite data is often very difficult since urban landscapes are composed of a very diverse assemblage of anthropogenic and natural materials (Mesev and Longley, 1999; Jensen, 2000). For example, a typical urban setting may have concrete, asphalt, plastic, metal, water, grass, shrubbery, trees, and soil. In fact, even when high spatial resolution data is used, conventional classification techniques have proven ineffective in urban areas (Gong and Howarth, 1992; Anys et al., 1998). Part of the problem with traditional urban classifications may be the single-pixel basis on which conventional classifications rely (Karathanassi et al., 2000). Indeed, the classification of remote sensing data relies on the assumption that the area being classified is composed of unique, internally homogeneous classes (Zhang and Foody, 1998). Therefore, significant progress in urban remote sensing will require novel methods to measure, model, and understand the dynamic nature of urban areas (Longley, 2002). One of these new methods may involve the use of image texture. This study examines the relationship between image texture and demographic variables in Terre Haute, IN, USA. Texture was derived from multispectral Advanced Spaceborne Thermal Emission Radiometer (ASTER) imagery.
In the past few decades, devastating earthquakes have caused high social and economic losses in cities. Earthquakes cannot be avoided, but the devastating impacts, especially fatalities, can be minimized through pre-event emergency response planning and preparedness. The development of emergency plans strongly relies on up-to-date population and inventory data. However, existing techniques for population data generation do not meet the requirements of many of today’s dynamic cities. In this context, the importance of remote sensing as a cutting-edge technology for data acquisition in urban areas is increasing. The present study analyzes the capacities and limitations of high resolution optical satellite imagery (IKONOS) for modeling population distribution in the district of Zeytinburnu in Istanbul, Turkey. The results show remote sensing to be an independent, up-to-date and area-wide data source. The use of remote sensing facilitates a mechanism to provide necessary quantitative information on urban morphology and population distribution in a fast and accurate way. The generated data do not have the quality of cadastral data sets but they meet the requirements of identifying bottlenecks, highly risky zones, etc. and can serve as a base for decision making.
Many organizations require accurate intermediate-scale landcover information for many applications, including modeling nutrient and pesticide runoff, understanding spatial patterns of biodiversity, land-use planning, and policy development. While many techniques have been successfully used to classify
land cover in relatively small regions, there are substantial obstacles in applying these methods to large, multiscene regions. The purpose of this study was to generate and evaluate a large region land-cover classification product using a multiple-layer land-characteristics database approach. To derive land-cover information, mosaicked Landsat thematic mapper (TM) scenes were analyzed in conjunction with digital
elevation data (and derived slope, aspect, and shaded relief. population census information, Defense Meteorological Satellite Program city lights data, prior land-use and landcover data, digital line graph data, and National Wetlands Inventory data. Both leaf-on and leaf-off TM data sets were analyzed. The study area was U.S. Federal Region 111, which includes the states of Pennsylvania, Virginia, Maryland, Delaware, and West Virginia.
The general procedure involved (1) generating mosaics of multiple scenes of leaves-on TM data using histogram equalization methods; (2) clustering mosaics into 100 spectral classes using unsupervised classification; (3) interpreting and labeling spectral classes into approximately 15 landcover
categories (analogous to Anderson Level 1 and 2 classes) using aerial photographs; (4) developing decisionmaking rules and models using from one to several ancillary data layers to resolve confusion in spectral classes that represented two or more targeted land-cover categories; and (5) incorporating data from other sources (for example, leaf-off TM data and National Wetlands Inventory data) to yield a final
land-cover product. Although standard accuracy assessments were not done, a series of consistency checks using available sources of land-cover information were conducted to evaluate the effectiveness of this approach for generating accurate land-cover information for large regions.
Texture is an important property of the images. Its inclusion in digital classification is known to improve the classification accuracy. In the present study, the texture features angular second moment, entropy and inverse difference moment were used to differentiate and classify forests affected by jhum (shifting cultivation) in north-eastern India. Large increases (11·1 per cent) in the classification accuracy were observed when texture and tone were used simultaneously. In general, the inverse difference moment was found to be more useful than the entropy. The angular second moment was not useful. The most accurate classification was achieved with a combination of the tone, the entropy and the inverse difference moment.
Spatial data are collected and represented as attributes of spatial objects embedded in a plane. Basis change is defined as the transfer of attributes from one set of objects to another. Methods of basis change for socioeconomic data are reviewed and are seen to differ in the assumptions made in each about underlying density surfaces. These methods are extended to more general cases, and an illustration is provided by using Californian data. The implementation of this framework within a geographical information system is discussed.
The authors are concerned with cross-estimation, or the estimation of data values for one set of reporting zones based on known values for another set. In essence this is the applied perspective on the modifiable areal unit problem, as cross-estimation is vitally important to people working with demographic or socioeconomic data in health, political redistricting, regional economic modeling and a number of other practically significant areas. It is also an information-limited problem - the more one knows about the area, the better the estimates. Information on some underlying factor might be used to improve the target zone estimates over the results of more naive techniques. -from Editors
This chapter reviews some of the problems involved in data integration, such as those of projection, digitizing error, scale and unit of analysis. Particular attention is paid to problems of integrating data for areal units. An account is given of work in progress at the North West Regional Research Laboratory (NWRRL) on improved methods for areal interpolation, one of the most important aspects of data integration. -from Authors
This paper conceives of residential segregation as a multidimensional phenomenon varying along five distinct axes of measurement:
evenness, exposure, concentration, centralization, and clustering. Twenty indices of segregation are surveyed and related
conceptually to one of the five dimensions. Using data from a large set of U.S. metropolitan areas, the indices are intercorrelated
and factor analyzed. Orthogonal and oblique rotations produce pattern matrices consistent with the postulated dimensional
structure. Based on the factor analyses and other information, one index was chosen to represent each of the five dimensions,
and these selections were confirmed with a principal components analysis. The paper recommends adopting these indices as standard
indicators in future studies of segregation.
Housing pressures in south-east England have intensified the interest in the urban fringe areas. Planners need to assess the requirements for new housing developments and amenity areas. Remotely-sensed data can provide planners with visually impacting material and a method of monitoring the urban fringe on medium-term time scales using change detection procedures. -from Authors
This paper presents a study of neural networks and version spaces for classification of remote sensing data. In the first network, precomputed textures based on the Spatial Grey Level Dependence (SGLD) method are fed to the net in conjunction with the spectral data. The second system is the sliding window network which uses all pixels in a small neighbourhood for classification of the central pixel. The third system is based on the candidate elimination implementation of the version space method for concept acquisition and is shown to achieve a performance similar to that of the neural systems but with faster training and symbolic rule generation.
This paper reviews and illustrates the use of satellite remote sensing in urban analysis. Using the example of conventional population census data used together with SPOT imagery for Strasbourg, France, it is shown that landscape quality can be characterized and mapped with a detail that is not possible using the census alone.
Land-cover classification using only remotely sensed spectral features does not provide accurate information on all urban-fringe classes. Four texture measures used in conjunction with LANDSAT spectral features were empirically evaluated to determine their utility in Level II and III land-cover mapping. The contrast and high-frequency measures improved land-cover classification at the urban fringe. However, the decision to use texture measures should be weighed carefully because they yield only a small, yet important, increment in absolute classification accuracy and entail additional expense for data preprocessing.
Geographic information systems (GIS) have been seen as an important enabling technology in the integration of satellite remote sensing and more conventional forms of land use information. The development of GIS has closely followed perceived application needs in physical geography and led to the implementation of numerous digital databases for land resources management. However, data relating to the human population and its socioeconomic characteristics have been poorly integrated with these advances for a number of reasons. Geographic information processing in the social sciences has tended to lag behind the development of systems for physical applications and, at a more fundamental level, there are important differences in the way in which socioeconomic phenomena have been conceptualized and modelled. Nevertheless, an understanding of the distribution and characteristics of human activity has an increasingly important role to play in many questions relating to the management of the physical environment. This paper seeks to demonstrate the application of some developments in socioeconomic analysis using GIS to integrate conventional physical resource data with socioeconomic information in the context of land use planning. These involve the modelling and analysis of population information in a raster GIS environment, incorporating contemporary geodemographic techniques and providing valuable information on the analysis of settlement and neighbourhood patterns and the distribution of residential, commercial and industrial activities.
Digital classification of Landsat data for use in natural resource inventory has produced mixed results. In attempts to improve classification, ancillary data, such as digitized maps and terrain (elevation) data, have been combined with Landsat data in various ways. These data have been used in (1) preclassification scene stratification; (2) postclassification class sorting; and (3) classification modification through increasing the number of observation channels, modifying prior probabilities, or adding a second stage to the classification. Preclassification stratification and postclassifications sorting are found to be efficient, but lacking in sophistication due to their reliance on deterministic decision rules. Classification modification through the simple addition of observations does not appear to improve classification results reliably; however, stratification of the sample used for training by the ancillary data does improve accuracy.
Segregation is commonly measured by means of an index of dissimilarity. A recent 'boundary modified' version of the index was formulated. It was based upon the concept that segregation is a separation created by spatial structure imposed upon the social space and thus interaction between racial groups is limited. The index takes into account one of the spatial etements—contiguity—but ignores the others. This paper argues that the length of the common boundary between two areal units and the shape of the areal units are important spatial components in determining segregation. Thus, a family of segregation indices is derived by incorporating these spatial components and can be applied to various spatial configurations. One of the indices possesses a distinctive property which is useful for comparing segregation levels in regions of various scales.
This study evaluates the increases in classification accuracy possible from satellite and airborne image texture processing and integration with ancillary topographic data for a moderate relief, boreal environment in Gros Morne National Park, eastern Canada. The texture measures angular second moment, entropy, and inverse difference moment were computed from spatial co-occurrence matrices in different orientations from SPOT multispectral linear array (MLA) and synthetic aperture radar (SAR) imagery. The general system of geomorphometry (elevation, slope, aspect, curvature, relief) was extracted from a co-registered digital elevation model (DEM). The three variables containing the greatest relative discriminatory power were MLA band 2, SAR angular second moment, and elevation. -from Authors
This paper investigates the incorporation of ancillary spatial data to improve the accuracy and specificity of a land-use classification from Landsat Thematic Mapper (TM) imagery for nonpoint source pollution modeling in a small urban area - the city of Beaver Dam, Wisconsin. A post-classification model was developed to identify and correct areas of confusion in the Landsat TM classification. Zoning and housing density data were used to modify the initial classification. Land-use classification accuracy improved and the number of identifiable classes increased. Additionally, confusion between classes that were commonly misclassified (for example, commercial and industrial areas) was reduced. -from Authors
Considers the nature of the modifiable areal unit problem. A survey is made of the prevailing ambivalent attitudes that geographers display and the general absence of any sense of verisimilitude is emphasized. A critical review is made of several alternative approaches to handling the problem.-from Authors
The quality of life (QOL) of the Athens-Clarke County of Georgia was assessed by integrating environmental variables extracted from Landsat thematic mapper data with socioeconomic variables obtained from the U.S. Census data. The environmental variables extracted are (1) land use/cover, particularly percentage of urban use; (2) normalized difference vegetation index (NDVI); and (3) apparent surface temperatures. The socioeconomic variables from the census data are (1) population density, (2) per capita income, (3) median home value, and (4) percentage of college graduates. NDVI, which shows strong negative correlation with “per cent of urban use” and “surface temperatures,” is in turn strongly correlated with per capita income, population density, and median home value. The environmental and socioeconomic variables are integrated by means of two approaches: (1) principal components analysis and (2) geographical-information-system overlay of ranked data layers. These two approaches constitute respectively an objective and a subjective assessment of the QOL of the study area. The results of these analyses indicate that satellite inwge data can contplentent census data in providing QOL assessment with an environmental perspective.
The information content of Landsat TM and MSS data was examined to assess the ability to digitally differentiate urban and near-urban land covers around Miami, FL. This examination included comparisons of unsupervised signature extractions for various cover types, training site statistics for intraclass and interclass separability, and band and band combination selection from an 11-band multisensor data set. The principal analytical tool used in this study was transformed divergence calculations. The TM digital data are typically more useful than the MSS data in the homogeneous near-urban land-covers and less useful in the heterogeneous urban areas.
Linking the results of remote sensing with production statistics to create land-use maps is problematic. This article presents a technique that integrates two significant data sets relating to agricultural land-use, namely, survey-based production statistics and vegetation growth profiles derived from satellite data. The article describes an application of this technique in which production data from an administrative survey are used to constrain the classification of NDVI profiles of land cover classes. Because the technique reallocates aggregated production statistics consistent with low-resolution imagery and with limited ground control information, it is a particularly cost-effective method of extensive land-use mapping.
Many efforts have been made to develop segregation indices that incorporate spatial interaction based on the contiguity concept. Contiguity refers to how similar the concentration of the subject of interest in one areal unit is to that in adjacent areal units. However, highly segregated situations are typically considered to be isolated sections or enclaves rather than smoothly formed peaks of concentration in space. Therefore, highly segregated enclaves may not exhibit contiguity. In this paper, a new index to measure the degree of clustering is developed and it is compared with the existing indices of concentration or segregation. The proposed clustering index ( I <sup>c</sup>) tends to give more weight to 'enclaveness' rather than contiguity alone. This may be a good property for those cases in which the primary concern of an investigator is the formation of enclaves of a socioeconomic subject, including minority populations, poverty, crime, epidemics, and mortgage red-lining. Additionally, its property of robustness to the citywide rate allows us to perform properly an intercity comparison of a given subject by index score even when the citywide rate varies significantly, unlike the other measures.
The problem of formal classification in urban morphological studies is addressed and a method of profiling the distinctive form of urban neighbourhoods is presented. The approach searches for regularity of pattern in tessellated surfaces of cities and constructs textural signatures that can distinguish between very fine differences in form. An experiment is reported that demonstrates the profiling power of these devices by using them as explanatory terms in a model that predicts residential density for 200 m<sup>2</sup> neighbourhoods. The wider application of image-based morphology measures is discussed.
The relationship between Jakub's recently introduced Distance-based Segregation Index and the Index of Dissimilarity is algebraically specified. The reasons why the distance-based index is substantially less dependent on the size of sub-areas are detailed, and the importance of this property is discussed. Some difficulties which the investigator is likely to encounter with the standardisation procedure are described.
CDATA91 Data Guide: 1991 Census of Population and Housing IDRISI for Windows User’s Guide, Version 2 Data integration: statistical methods for transferring data between zonal systems
- Canberra R Eastman
Australian Bureau of Statistics (1993) CDATA91 Data Guide: 1991 Census of Population and Housing, Canberra. Eastman R. (1997) IDRISI for Windows User’s Guide, Version 2.0 (Worcester, MA: Graduate School of Geography, Clark University). Flowerdew, R. and M. Green (1991) Data integration: statistical methods for transferring data between zonal systems, In Handling Geographical Information: Methodology and Applications, edited by I. Masser and M
Mapping population data from zone centroid locations. Transactions of the Institute of British Geographers
- D Martin
Martin D. (1989) Mapping population data from zone centroid locations. Transactions of the Institute of British
Geographers, 14, pp. 90-97.
RS-GIS: spatial distributions from remote imagery
- V Mesev
- P Longley
- M Batty
Mesev, V., P. Longley and M. Batty (1996) RS-GIS: spatial distributions from remote imagery. In Spatial
Analysis: Modelling in a GIS Environment, edited by P. Longley and M. Batty (London: GeoInformation
International).
CDATA91 Data Guide: 1991 Census of Population and Housing
Australian Bureau of Statistics (1993) CDATA91 Data Guide: 1991 Census of Population and Housing,
Canberra.
Use of population density to improve classification accuracy in remotely sensed images of urban areas
- G J Sadler
- M J Barnsley
Sadler, G.J. and M.J. Barnsley (1990) Use of population density to improve classification accuracy in remotely
sensed images of urban areas. EGIS (90), Amsterdam, The Netherlands.
Hornsby Council land Information System -Cadastral Information, Hornsby Council Central Library
- Council Hornsby Shire
Hornsby Shire Council (1992) Hornsby Council land Information System -Cadastral Information, Hornsby
Council Central Library, Sydney, Australia.