Nonconvex Compressive Sensing Optimal Reconstruction Method Based on Sketch Representation and Structured Clustering

A compressed sensing and structured technology, applied in the field of image processing, which can solve problems such as unfavorable real-time applications, slow reconstruction speed, and lack of accuracy and robustness of reconstruction.

CN107492129BActive Publication Date: 2021-01-19XIDIAN UNIV
4 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Publication Date
2021-01-19

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention discloses a non-convex compressive sensing optimization reconstruction method based on sketch representation and structured clustering, which mainly solves the problem of inaccurate reconstruction of compressive sensing images at low sampling rates. The realization process is as follows: according to the sketch map of the image , defines sketchable blocks and non-sketchable blocks, where non-sketchable blocks include smooth blocks and texture blocks, and sketchable blocks include unidirectional and multi-directional blocks; unidirectional blocks use clustering based on sketch direction guidance; multi-directional blocks use Clustering based on directional distribution features; smooth blocks and texture blocks adopt grayscale clustering; multi-measurement vector observations are performed for each type of image block; when reconstructing, according to the multi-measurement matrix, category index and The direction information adopts the particle swarm algorithm based on intersection and atomic direction constraints to obtain the final reconstructed image. Compared with the TS_RS and NR_DG methods, the reconstructed image has high quality and good robustness, and can be used for the reconstruction of natural images.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention belongs to the technical field of image processing, and in particular relates to a non-convex compressive sensing optimization reconstruction method based on sketch representation and structured clustering, which can be used to reconstruct natural images. Background technique

[0002] In recent years, a new data theory compressive sensing CS has emerged in the field of signal processing. This theory realizes compression while collecting data, breaks through the limitations of the traditional Nyquist sampling theorem, and brings new advantages to data collection technology. The revolutionary changes make the theory have broad application prospects in compressed imaging systems, military cryptography, wireless sensing and other fields. Compressed sensing theory mainly includes three aspects: sparse representation of signal, observation of signal and reconstruction of signal. Designing effective observation and reconstruction methods is an ...

Examples

Embodiment

[0153] 1, emulation condition: the emulation of the present invention is at windows 7, SPI, CPU Intel (R) Core (TM) i5-3470, basic frequency 3.20GHz, software platform is to run on Matlab R2011b, what emulation selects is four frames of 512 * 512 Standard test natural images Lena, Barbara, Boat, block size.

[0154]2. Simulation content and results:

[0155] Simulation 1:

[0156] Under the condition that the sampling rate is 20%, the Barbara image is reconstructed respectively with the method of the present invention and the existing method, and the simulation results are shown in Fig. image 3 As shown, among them, image 3 (a) is the original picture of Barbara, image 3 (b) for image 3 Partial enlarged view of (a), image 3 (c) is the reconstruction map obtained by the two-stage reconstruction method (TS_RS), image 3 (d) for image 3 Partial enlarged view of (c), image 3 (e) is the reconstruction map obtained by direction-guided reconstruction (NR_DG), image 3...