Comparison between pixel- and object-based image classification of a tropical landscape using Système Pour l’Observation de la Terre-5 imagery

Abstract

Abstract. Based on the Système Pour l'Observation de la Terre-5 imagery, two main techniques of classifying land-use categories in a tropical landscape are compared using two supervised algorithms: maximum likelihood classifier (MLC) and K-nearest neighbor object-based classifier. Nine combinations of scale level (SL10, SL30, and SL50) and the nearest neighbor (NN3, NN5, and NN7) are investigated in the object-based classification. Accuracy assessment is performed using two main disagreement components, i.e., quantity disagreement and allocation disagreement. The MLC results in a higher total disagreement in total landscape as compared with object-based image classification. The SL30-NN5 object-based classifier reduces allocation error by 250% as compared with the MLC. Therefore, this classifier shows a higher performance in land-use classification of the Langat basin. © The Authors. Published by SPIE under a Creative Commons Attribution 3.0 Unported License. Distribution or reproduction of this work in whole or in part requires full attribution of the original publication, including its DOI.

An Evaluation of Pixel-based and Object-based Classification Methods for Land Use Land Cover Analysis Using Geoinformatic Techniques

“…The maximum likelihood (ML) classifier supervised classification algorithm was employed in this investigation. The training data is used to estimate the probabilities [20] that a pixel belongs to a certain class which based on a subsequent probability of membership and dimensions equal to numerous bands in the source image [3]. The training signatures (pixels) for each class were chosen to represent the majority of the class's spectral characteristics over the entire image.…”

“…The LULC data provides a panoramic view of the landscape's characteristics and facilitates decision-making on interconnected parts of most land-based processes [3]. It can be obtained from satellite images using remote sensing image classification [4,5].…”

Synergistic Use of Radar and Optical Satellite Data for Improved Monsoon Cropland Mapping in India

“…Further image classification was performed on S1 ( Figure 2b), and S1+S2 combined (Figure 2c), using pixel-based machine learning classifier (random forest) on GEE. We have used pixel-based classifier instead of object-based classifier for large monsoon cropland mapping, as the latter requires high computation time and has complicated intermediate steps including the segmentation where specific parameter tuning is needed [32,63,64]. Even though object-based classifiers might improve the classification accuracy in some landscapes, this performance improvement is not always evident in complex heterogeneous landscapes such as the one showed in this study.…”

Comparison between Svm and MLC to Map the Land Cover Using Sentinel 2 Data In Babel Governorate – Iraq

“…Performing well at large spatial extents in requiring fewer sets of training samples due to the fewer classifications performed. Some advantages to MLC is: easy to use, well defined, and has been accessible to researchers for many years (24). The high accessibility of Maximum Likelihood Classifier (MLC) gives a benefit of being well known by different researchers and used in different studies, allowing new results of MLC to be compared against many studies.…”

“…Mixed pixels contain multiple land-cover classes within the single pixel, which results in the pixel value not fully representing any of the desired classes (24) (17). Increasing the resolution of the image can reduce the number and impact of mixed pixels on an image (24). Over the past years, (SVM) Support Vector Machine capabilities have been shown to be a good option as data classifier for remotely sensed imaging.…”