Multispectral Image Change Detection Based on Single-Band Slow Feature Analysis

• Published in: Remote sensing (Basel, Switzerland) Vol. 13; no. 15; p. 2969
• Main Authors: He, Youxi, Jia, Zhenhong, Yang, Jie, Kasabov, Nikola K.
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 01.08.2021
• Subjects: Accuracy; Algorithms; Change detection; Cluster analysis; Clustering; Deep learning; Dictionaries; Euclidean geometry; filters; Gray scale; image analysis; Image filters; Machine learning; multispectral imagery; multispectral remote sensing image; Neural networks; Noise; Principal components analysis; problem solving; Radiation; Remote sensing; slow feature analysis; Vector quantization
• DOI: 10.3390/rs13152969
• ISSN: 2072-4292

Due to differences in external imaging conditions, multispectral images taken at different periods are subject to radiation differences, which severely affect the detection accuracy. To solve this problem, a modified algorithm based on slow feature analysis is proposed for multispectral image change detection. First, single-band slow feature analysis is performed to process bitemporal multispectral images band by band. In this way, the differences between unchanged pixels in each pair of single-band images can be sufficiently suppressed to obtain multiple feature-difference images containing real change information. Then, the feature-difference images of each band are fused into a grayscale distance image using the Euclidean distance. After Gaussian filtering of the grayscale distance image, false detection points can be further reduced. Finally, the k-means clustering method is performed on the filtered grayscale distance image to obtain the binary change map. Experiments reveal that our proposed algorithm is less affected by radiation differences and has obvious advantages in time complexity and detection accuracy.