Assessing the performance of the multi-morphological profiles in urban land cover mapping using pixel based classifiers and very high resolution satellite imagery (original) (raw)
Related papers
Per-pixel Classification of High Spatial Resolution Satellite Imagery for Urban Land-cover Mapping
Photogrammetric Engineering & Remote Sensing, 2008
Commercial high spatial resolution satellite data now provide a synoptic and consistent source of digital imagery with detail comparable to that of aerial photography. In the work described here, per-pixel classification, image fusion, and GIS-based map refinement techniques were tailored to pan-sharpened 0.61 m QuickBird imagery to develop a six-category urban land-cover map with 89.3 percent overall accuracy (ϭ 0.87). The study area was a rapidly developing 71.5 km 2 part of suburban Raleigh, North Carolina, U.S.A., within the Neuse River basin. "Edge pixels" were a source of classification error as was spectral overlap between bare soil and impervious surfaces and among vegetated cover types. Shadows were not a significant source of classification error. These findings demonstrate that conventional spectral-based classification methods can be used to generate highly accurate maps of urban landscapes using high spatial resolution imagery.
Urban Classification from Aerial and Satellite Images
Journal of Applied Engineering Sciences, 2020
When talking about land cover, we need to find a proper way to extract information from aerial or satellite images. In the field of photogrammetry, aerial images are generally acquired by optical sensors that deliver images in four bands (red, green, blue and near-infrared). Recent researches in this field demonstrated that for the image classification process is still place for improvement. From satellites are obtained multispectral images with more bands (e.g. Landsat 7/8 has 36 spectral bands). This paper will present the differences between these two types of images and the classification results using support-vector machine and maximum likelihood classifier. For the aerial and the satellite images we used different sets of classification classes and the two methods mentioned above to highlight the importance of choosing the classes and the classification method.
International Journal of Remote Sensing, 2012
Detailed, up-to-date information on intra-urban land cover is important for urban planning and management. Differentiation between permeable and impermeable land, for instance, provides data for surface run-off estimates and flood prevention, whereas identification of vegetated areas enables studies of urban micro-climates. In place of maps, high-resolution images, such as those from the satellites IKONOS II, Quickbird, Orbview and WorldView II, can be used after processing. Object-based image analysis (OBIA) is a well-established method for classifying high-resolution images of urban areas. Despite the large number of previous studies of OBIA in the context of intra-urban analysis, there are many issues in this area that are still open to discussion and resolution. Intra-urban analysis using OBIA can be lengthy and complex because of the processing difficulties related to image segmentation, the large number of object attributes to be resolved and the many different methods needed to classify various image objects. To overcome these issues, we performed an experiment consisting of land-cover mapping based on an OBIA approach using an IKONOS II image of a southern sector of São José dos Campos city (covering an area of 12 km2 with 50 neighbourhoods), which is located in São Paulo State in south-eastern Brazil. This area contains various occupation and land-use patterns, and it therefore contains a wide range of intra-urban targets. To generate the land-cover map, we proposed an OBIA-based processing framework that combines multi-resolution segmentation, data mining and hierarchical network techniques. The intra-urban land-cover map was then evaluated through an object-based error matrix, and classification accuracy indices were obtained. The final classification, with 11 classes, achieved a global accuracy of 71.91%.
Photogrammetric Engineering and Remote Sensing, 2013
Although spatial measures such as texture and shape extracted from very high resolution imagery (VHR) have been successfully employed in pixel-based classifications, the effectiveness of such measures in classification mainly depends on the optimal window size in which spatial measures are calculated. However, an optimal window size is usually subjective and varies for different image and different land cover types. Multiresolution segmentation of object-based image analysis, on the other hand, results objects with different size and shape, which are meaningful and better represent the real size and shape of land cover types. This paper introduces a new approach to land cover classification which benefits from both pixel-based and object-based image analyses. The VHR image is firstly segmented into different levels of segmentations. For each level, one set of spectral measures and two sets of spatial measures, texture and morphology, are extracted and then stacked to the original bands of VHR image forming a several-band image. To determine the contribution of each set of measures in separating urban land cover classes, the separability distance for all class pairs are calculated based on Bhattacharryya distance for each set of measures (i.e. spectral, texture and morphology). A pixel-based maximum likelihood classification is then applied to each set of bands. Results show that adding either texture or morphology to the original bands of VHR image has almost the same effect in increasing the overall classification accuracy. Furthermore, the classification accuracy of buildings and roads increases significantly by incorporation of spatial measures in classification procedure.
2012
Cities are centers of human activity and more than half of the world’s population live in metropolitan areas. Urban areas are characterized by a large variety of artificial and natural surface materials influencing ecological, climatic and energetic conditions. With advent of new sensors in remote sensing fields that capture the data in high spatial and spectral resolution we are able to classify the urban area accurately. The main goal of this study is, comparison between different pixels based classification methods such as Maximum likelihood, Minimum distance, spectral angel mapper and Support Vector Machine (SVM).Thus to apply these methods we use pansharpening image of worldview 2 from Kuala Lumpur, Malaysia with 0.6 meter spatial resolution and 8 spectral bands. The results show that SVM method more accurate than other methods in classify the urban area with 72% overall accuracy with Kappa coefficient 0.65.However the very high resolution data shows the good potential for clas...
South African Journal of Geomatics
The identification, extraction, classification and mapping of detailed, but reliable Land Use or Land Cover (LULC) data play an increasingly important role in informed decision-making whether employed in urban planning and civil engineering, intensive agriculture, the natural and environmental sciences, for example. One way of extracting LULC information is through the use of algorithms that classify multispectral satellite images according to the required standard and user legend. The meaningful classification of heterogeneous urban and city landscapes however remains challenging and is performed using semi-automated pixel-based, object-based, or hybrid classification workflows. With the prevailing remote sensing technologies enabling professionals to integrate multidimensional data from various sources to improve the quality of LULC classification nowadays, it negated the dependency on (multi)spectral information alone. This study sought to explore how successful a single-acquisition pansharpened SPOT 6 image can be deconstructed into obtaining primary and secondary LULC classes. This was achieved using a comparison of the pixel-based versus segmentation-based classifiers, performed over Soshanguve Township in South Africa. The study further assessed the effect of integrating LiDAR derived 3D land surface data into both classification processes. A supervised Maximum Likelihood classifier was executed for the pixel-based routine, while the ERDAS IMAGINE Objective tool was used for the segmentationbased approach. A total of nine LULC classes were successfully identified from the classification. The results showed that the segmentation-based approach outperformed the pixel-based approach, yet when integrating height information both segmentation and pixel-based overall accuracies
Land use land cover analysis with pixel-based classification approach
Indonesian Journal of Electrical Engineering and Computer Science, 2019
Rapid development in certain urban area will affect its natural features. Therefore, it is important to identify and determine the changes occur for further analysis and future development planning. This process will influence several factors such as area development, environmental issues and human social activities. The selection of remote sensing data and method will derive the accurate land use land cover maps. This research study accessed the classification accuracy of different classifier approach for land use land cover classification in urban area. The objective of this paper is to compare the accuracy of the classification for each technique used. The study was conducted in a highly urbanized area in Kuala Lumpur, Malaysia. The dataset used for this study is the multi temporal LANDSAT satellite imageries for the year of 2001,2006,2011 and 2016. The pre-processing and analysis of the dataset has been done using software ENVI 5.3. Five land use classes (Urban/built up area, Fo...
Using Pixel-Based and Object-Based Methods to Classify Urban Hyperspectral Features
Geodesy and cartography, 2016
Object-based image analysis methods have been developed recently. They have since become a very active research topic in the remote sensing community. This is mainly because the researchers have begun to study the spatial structures within the data. In contrast, pixel-based methods only use the spectral content of data. To evaluate the applicability of object-based image analysis methods for land-cover information extraction from hyperspectral data, a comprehensive comparative analysis was performed. In this study, six supervised classification methods were selected from pixel-based category, including the maximum likelihood (ML), fisher linear likelihood (FLL), support vector machine (SVM), binary encoding (BE), spectral angle mapper (SAM) and spectral information divergence (SID). The classifiers were conducted on several features extracted from original spectral bands in order to avoid the problem of the Hughes phenomenon, and obtain a sufficient number of training samples. Three...
Tsukuba geoenvironmental …, 2007
Although several techniques to extract the land uses from remotely sensed data have been evolving, mapping urban landscape with enough accuracy is not completely achieved. This paper aims to evaluate image classification methods for mapping the urban landscape of a fast growing city in the Tokyo metropolitan fringe using Advanced Land Observing Satellite (ALOS) data. Three image classification methods: unsupervised, supervised and fuzzy supervised were evaluated. An AVNIR-2 sensor image of ALOS satellite covering Tsukuba city was used for the study. Field survey data including high resolution satellite image and aerial photographs were used for scheming land use types, selecting training samples and assessing accuracies of the classification results. Seven types of land uses: forested land; lawn/ grass; paddy field; dry farmland/exposed field; facility/ industry; residence/parking/road/upland bare field and water were extracted using the methods. Error matrix and Kappa index were computed to measure the map accuracy. The fuzzy supervised method improved the mapping results showing highest overall accuracy of 87.7% as compared to supervised and unsupervised methods. The fuzzy method effectively dealt with the mixed pixels that appeared in the residential area. The study also revealed that the image classification method greatly influences the spatial statistics of land use types.
Classification of CBERS-02B high resolution image using morphological features for urban areas
2012 Second International Workshop on Earth Observation and Remote Sensing Applications, 2012
Urban landscapes represent one of the most challenging areas for remote sensing analysis due to high spatial and spectral diversities of surface materials involved. High Resolution images (HR, better than 5-m spatial resolution) have a potential for detailed and accurate mapping of urban environment. The objective of this study is to analyze the effectiveness of multi-scale morphological features in the purpose of classifying urban landscapes with panchromatic HR images. The experiment is performed using two CBERS HR scenes with urban landscapes characterized by different architectural styles, namely an apartment block and a peri-urban village surrounding Beijing City. Seven types of morphological features including opening (O), closing (C), opening by reconstruction(OR), closing by reconstruction(CR), opening by top-hat(OTH), closing by top hat(CTH) and derivative morphological profile(DMP) are assessed. A support vector machine classifier was also employed to handle the considerable amount of morphological features. The classification results show that with multi-scale morphological features it is possible to discriminate surfaces with mixed spectral characters such as roads, parking lots, and tents due to their different textures for each scene. According to the validation results, the overall accuracy can be improved from 50% with single band HR data to 80.2% and 76.3% respectively for each urban scene using HR-OC-DMP morphological sets. The classification of residential buildings with similar textual character but different gray scales is improved a lot with the supplement of OC sets. The integration of DMP and OC sets benefits the differentiation of bare soil and roads. The mixed sets combining simple, reconstruction and DMP filters provide the best performance.