Exploration of the influence of ambiguity pixels on image classification reliability

Ambiguity pixels and unreliable classification are easily generated in remote sensing images (RSIs) due to the complexity of landscape sand imaging conditions. At present, few studies focus on the quantification, influence, and propagation mechanism of ambiguity pixels in RSI classification. To investigate these problems, this paper extracts the ambiguity pixels in the image, and explores their influence on classification reliability. First, the pixels of an RSI are classified into ambiguity and unambiguity classes on the basis of uncertainty. Second, the reliability of classification results on ambiguity and unambiguity pixels is evaluated and analyzed using defined reliability indices. Finally, three experiments are designed to reveal the influence of ambiguity pixels on the reliability of RSI classification. Experimental results show that the indicator values of the unambiguity pixels are much higher than those of the ambiguity pixels, illustrating the effects of ambiguity pixels on the reliability of RSI classification results.

[1]  Ling Han,et al.  Standardized object-based dual CNNs for very high-resolution remote sensing image classification and standardization combination effect analysis , 2020 .

[2]  Manchun Li,et al.  Using convolutional neural network to identify irregular segmentation objects from very high-resolution remote sensing imagery , 2018 .

[3]  Lei Lei,et al.  Approach of information fusion and classification by SVM and DS evidence theory , 2013 .

[4]  Shao Wei.  Study on Rule Set Construction and Application to Land Use Classification , 2012 .

[5]  Wenzhong Shi,et al.  A Reliability-Based Multi-Algorithm Fusion Technique in Detecting Changes in Land Cover , 2013, Remote. Sens..

[6]  Shihong Du,et al.  Spectral–Spatial Feature Extraction for Hyperspectral Image Classification: A Dimension Reduction and Deep Learning Approach , 2016, IEEE Transactions on Geoscience and Remote Sensing.

[7]  Shi Lei,et al.  Environment as the Stage for Economic Actors , 2007 .

[8]  Andrew P Brna,et al.  Uncertainty-based modulation for lifelong learning , 2019, Neural Networks.

[9]  Kun Zhu,et al.  An object-based supervised classification framework for very-high-resolution remote sensing images using convolutional neural networks , 2018 .

[10]  Stefan Adriaensen,et al.  Remote Sensing Letters , 2013 .

[11]  Lei Shi,et al.  Feature Selection for Object-Based Classification of High-Resolution Remote Sensing Images Based on the Combination of a Genetic Algorithm and Tabu Search , 2018, Comput. Intell. Neurosci..

[12]  Zhou Yang,et al.  Scene classification of remote sensing image based on deep network and multi-scale features fusion , 2018, Optik.

[13]  Siamak Khorram,et al.  High-resolution land cover change detection based on fuzzy uncertainty analysis and change reasoning , 2010 .

[14]  Tarald O. Kvålseth,et al.  A Coefficient of Agreement for Nominal Scales: An Asymmetric Version of Kappa , 1991 .

[15]  Li Zhengyan,et al.  Quantum immune optimization algorithm based on cloud model , 2011 .

[16]  Feng Guoh,et al.  Parameter optimizing for Support Vector Machines classification , 2011 .

[17]  Saeid Nahavandi,et al.  A Classifier Graph Based Recurring Concept Detection and Prediction Approach , 2018, Comput. Intell. Neurosci..

[18]  Alfonso Fernández-Manso,et al.  Spectral unmixing , 2012 .

[19]  Wang Bin,et al.  Research progress of deep learning in classification and recognition of remote sensing images , 2019 .