Normalization in Unsupervised Segmentation Parameter Optimization: A Solution Based on Local Regression Trend Analysis

Georganos, Stefanos; Lennert, Moritz; Grippa, Taïs; Vanhuysse, Sabine; Johnson, Brian Alan; Wolff, Éléonore

doi:10.3390/rs10020222

Cited by 20 publications

(23 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The MRS algorithm has three free parameters: scale parameter (SP), color/shape weights, and smoothness/compactness weights. The SP is the most important parameter because it implicitly governs the average size of image segments, and setting it appropriately is critical for achieving accurate segmentation and classification results [46][47][48]. This parameter is conventionally set through a time-consuming and subjective trial-and-error process.…”

Section: Image Segmentation and Feature Extractionmentioning

confidence: 99%

Comparing Deep Neural Networks, Ensemble Classifiers, and Support Vector Machine Algorithms for Object-Based Urban Land Use/Land Cover Classification

2019

View full text Add to dashboard Cite

With the advent of high-spatial resolution (HSR) satellite imagery, urban land use/land cover (LULC) mapping has become one of the most popular applications in remote sensing. Due to the importance of context information (e.g., size/shape/texture) for classifying urban LULC features, Geographic Object-Based Image Analysis (GEOBIA) techniques are commonly employed for mapping urban areas. Regardless of adopting a pixel- or object-based framework, the selection of a suitable classifier is of critical importance for urban mapping. The popularity of deep learning (DL) (or deep neural networks (DNNs)) for image classification has recently skyrocketed, but it is still arguable if, or to what extent, DL methods can outperform other state-of-the art ensemble and/or Support Vector Machines (SVM) algorithms in the context of urban LULC classification using GEOBIA. In this study, we carried out an experimental comparison among different architectures of DNNs (i.e., regular deep multilayer perceptron (MLP), regular autoencoder (RAE), sparse, autoencoder (SAE), variational autoencoder (AE), convolutional neural networks (CNN)), common ensemble algorithms (Random Forests (RF), Bagging Trees (BT), Gradient Boosting Trees (GB), and Extreme Gradient Boosting (XGB)), and SVM to investigate their potential for urban mapping using a GEOBIA approach. We tested the classifiers on two RS images (with spatial resolutions of 30 cm and 50 cm). Based on our experiments, we drew three main conclusions: First, we found that the MLP model was the most accurate classifier. Second, unsupervised pretraining with the use of autoencoders led to no improvement in the classification result. In addition, the small difference in the classification accuracies of MLP from those of other models like SVM, GB, and XGB classifiers demonstrated that other state-of-the-art machine learning classifiers are still versatile enough to handle mapping of complex landscapes. Finally, the experiments showed that the integration of CNN and GEOBIA could not lead to more accurate results than the other classifiers applied.

show abstract

Section: Image Segmentation and Feature Extractionmentioning

confidence: 99%

Comparing Deep Neural Networks, Ensemble Classifiers, and Support Vector Machine Algorithms for Object-Based Urban Land Use/Land Cover Classification

2019

View full text Add to dashboard Cite

show abstract

“…In the GRASS implementation, the TP ranges between 0 to 1, with 0 leading to the situation where each pixel represents a segment, while 1 unifies all image pixels in one object. As Böck et al [52] pointed out, the USPO metrics are sensitive to the range of candidate segmentations used as input, so we empirically found a range that corresponded to cases of evident over-and under-segmentations to be used as minimum and maximum possible values, as commonly done in similar studies [18,53]. Thus, we evaluated 27 different segmentations starting with a TP of 4 and finishing at a TP of 31, guided by an incrementing step value of 1, as in previous studies, [54].…”

Section: Segmentation and Unsupervised Segmentation Parameter Optimizmentioning

confidence: 73%

“…where for Equation 1, n is the number of segments, v i is the variance and a i the area for each segment, while in Equation 2, n is the number of segments, z i = x i − x, x is the mean value of segment x, M = ∑ n i=1 ∑ n j=1 w ij and w ij is the element of the matrix of spatial proximity M, which indicates the spatial connectivity for segments i and j [52,53].…”

Section: Segmentation and Unsupervised Segmentation Parameter Optimizmentioning

confidence: 99%

See 1 more Smart Citation

Scale Matters: Spatially Partitioned Unsupervised Segmentation Parameter Optimization for Large and Heterogeneous Satellite Images

et al. 2018

Self Cite

View full text Add to dashboard Cite

To classify Very-High-Resolution (VHR) imagery, Geographic Object Based Image Analysis (GEOBIA) is the most popular method used to produce high quality Land-Use/Land-Cover maps. A crucial step in GEOBIA is the appropriate parametrization of the segmentation algorithm prior to the classification. However, little effort has been made to automatically optimize GEOBIA algorithms in an unsupervised and spatially meaningful manner. So far, most Unsupervised Segmentation Parameter Optimization (USPO) techniques, assume spatial stationarity for the whole study area extent. This can be questionable, particularly for applications in geographically large and heterogeneous urban areas. In this study, we employed a novel framework named Spatially Partitioned Unsupervised Segmentation Parameter Optimization (SPUSPO), which optimizes segmentation parameters locally rather than globally, for the Sub-Saharan African city of Ouagadougou, Burkina Faso, using WorldView-3 imagery (607 km2). The results showed that there exists significant spatial variation in the optimal segmentation parameters suggested by USPO across the whole scene, which follows landscape patterns—mainly of the various built-up and vegetation types. The most appropriate automatic spatial partitioning method from the investigated techniques, was an edge-detection cutline algorithm, which achieved higher classification accuracy than a global optimization, better predicted built-up regions, and did not suffer from edge effects. The overall classification accuracy using SPUSPO was 90.5%, whilst the accuracy from undertaking a traditional USPO approach was 89.5%. The differences between them were statistically significant (p < 0.05) based on a McNemar’s test of similarity. Our methods were validated further by employing a segmentation goodness metric, Area Fit Index (AFI)on building objects across Ouagadougou, which suggested that a global USPO was more over-segmented than our local approach. The mean AFI values for SPUSPO and USPO were 0.28 and 0.36, respectively. Finally, the processing was carried out using the open-source software GRASS GIS, due to its efficiency in raster-based applications.

show abstract

“…To quantitatively evaluate image segmentation results, supervised [41,42] and unsupervised methods [43,44] are proposed. The supervised methods are more objective for evaluating the image segmentation quality [41].…”

Section: Segmentation Evaluationmentioning

confidence: 99%

Context-Enabled Extraction of Large-Scale Urban Functional Zones from Very-High-Resolution Images: A Multiscale Segmentation Approach

Liu

Zhang

2019

Remote Sensing

View full text Add to dashboard Cite

Urban functional-zone (UFZ) analysis has been widely used in many applications, including urban environment evaluation, and urban planning and management. How to extract UFZs’ spatial units which delineates UFZs’ boundaries is fundamental to urban applications, but it is still unresolved. In this study, an automatic, context-enabled multiscale image segmentation method is proposed for extracting spatial units of UFZs from very-high-resolution satellite images. First, a window independent context feature is calculated to measure context information in the form of geographic nearest-neighbor distance from a pixel to different image classes. Second, a scale-adaptive approach is proposed to determine appropriate scales for each UFZ in terms of its context information and generate the initial UFZs. Finally, the graph cuts algorithm is improved to optimize the initial UFZs. Two datasets including WorldView-2 image in Beijing and GaoFen-2 image in Nanchang are used to evaluate the proposed method. The results indicate that the proposed method can generate better results from very-high-resolution satellite images than widely used approaches like image tiles and road blocks in representing UFZs. In addition, the proposed method outperforms existing methods in both segmentation quality and running time. Therefore, the proposed method appears to be promising and practical for segmenting large-scale UFZs.

show abstract

Normalization in Unsupervised Segmentation Parameter Optimization: A Solution Based on Local Regression Trend Analysis

Cited by 20 publications

References 44 publications

Comparing Deep Neural Networks, Ensemble Classifiers, and Support Vector Machine Algorithms for Object-Based Urban Land Use/Land Cover Classification

Comparing Deep Neural Networks, Ensemble Classifiers, and Support Vector Machine Algorithms for Object-Based Urban Land Use/Land Cover Classification

Scale Matters: Spatially Partitioned Unsupervised Segmentation Parameter Optimization for Large and Heterogeneous Satellite Images

Context-Enabled Extraction of Large-Scale Urban Functional Zones from Very-High-Resolution Images: A Multiscale Segmentation Approach

Contact Info

Product

Resources

About