Multispectral single-pixel imaging using predecoded virtual patterns

doi:10.1016/j.optlaseng.2024.108099

Multispectral single-pixel imaging using predecoded virtual patterns

2024-05-01

1

PRE

AI

DOI：10.1016/j.optlaseng.2024.108099

原文链接

原文求助

摘要

En 中文

Multispectral single -pixel imaging has garnered increasing attention in recent years due to its capacity for acquiring multispectral object information. Multispectral multiplexing structure illumination is a typical modulation-based encoding method that employs a single -pixel detector to synchronously collect and obtain multispectral object information. However, in these methods, it is necessary to first recover the mixed multispectral image using iterative solution algorithms. Subsequently, decoding processing is employed to extract each component of multispectral object information. Based on the reciprocity between mixed multispectral image decoding and iterative solution patterns decoding, this paper proposes a scheme that acquires predecoded patterns in advance to directly obtain each component of multispectral information. This scheme incorporates nonexperimental predecoded virtual patterns into an iterative solution algorithm, enabling the direct acquisition of multispectral object information. This research provides a comprehensive exposition of the proposed methodology and substantiates its efficacy through experimental investigations. The proposed method is fundamentally different from existing methods that require mixed multispectral images to be obtained for postdecoding processing. It is a direct predecoding-based method that enhances the efficiency of multispectral imaging.

Keyword:

Single -pixel imaging

Decoding virtual patterns

Multispectral information

Multispectral single-pixel imaging using predecoded virtual patterns

摘要

期刊

机构