Solder joint imagery compressing and recovery based on compressive sensing

Abstract

Purpose

The purpose of this paper is to develop an improved compressive sensing algorithm for solder joint imagery compressing and recovery. The improved algorithm can improve the performance in terms of peak signal to noise ratio (PSNR) of solder joint imagery recovery.

Design/methodology/approach

Unlike the traditional method, at first, the image was transformed into a sparse signal by discrete cosine transform; then the solder joint image was divided into blocks, and each image block was transformed into a one-dimensional data vector. At last, a block compressive sampling matching pursuit was proposed, and the proposed algorithm with different block sizes was used in recovering the solder joint imagery.

Findings

The experiments showed that the proposed algorithm could achieve the best results on PSNR when compared to other methods such as the orthogonal matching pursuit algorithm, greedy basis pursuit algorithm, subspace pursuit algorithm and compressive sampling matching pursuit algorithm. When the block size was 16 × 16, the proposed algorithm could obtain better results than when the block size was 8 × 8 and 4 × 4.

Practical implications

The paper provides a methodology for solder joint imagery compressing and recovery, and the proposed algorithm can also be used in other image compressing and recovery applications.

Originality/value

According to the compressed sensing (CS) theory, a sparse or compressible signal can be represented by a fewer number of bases than those required by the Nyquist theorem. The findings provide fundamental guidelines to improve performance in image compressing and recovery based on compressive sensing.

Citation

Zhao, H., Wang, Y., Qiao, Z. and Fu, B. (2014), "Solder joint imagery compressing and recovery based on compressive sensing", Soldering & Surface Mount Technology, Vol. 26 No. 3, pp. 129-138. https://doi.org/10.1108/SSMT-09-2013-0024