Abstract
Based on the advantages of a non-subsampled shearlet transform (NSST) in image processing and the characteristics of remote sensing imagery, NSST was applied to enhance blurred images. In the NSST transform domain, directional information measurement can highlight textural features of an image edge and reduce image noise. Therefore, NSST was applied to the detailed enhancement of high-frequency sub-band coefficients. Based on the characteristics of a low-frequency image, the retinex method was used to enhance low-frequency images. Then, an NSST inverse transformation was performed on the enhanced low- and high-frequency coefficients to obtain an enhanced image. Computer simulation experiments showed that when compared with a traditional image enhancement strategy, the method proposed in this paper can enrich the details of the image and enhance the visual effect of the image. Compared with other algorithms listed in this paper, the brightness, contrast, edge strength, and information entropy of the enhanced image by this method are improved. In addition, in the experiment of noisy images, various objective evaluation indices show that the method in this paper enhances the image with the least noise information, which further indicates that the method can suppress noise while improving the image quality, and has a certain level of effectiveness and practicability.
Highlights
Various factors, such as the imaging environment, system equipment, and transmission medium, often affect the image acquisition process and cause the quality of an image to degrade by varying degrees
The enhanced image brightness of methods 2 and 3 as well as the method used in this paper is better
Based on the advantages of non-subsampled shearlet transform (NSST) in the field of image processing and the working principle of image enhancement, NSST was applied to the field of image enhancement
Summary
Various factors, such as the imaging environment, system equipment, and transmission medium, often affect the image acquisition process and cause the quality of an image to degrade by varying degrees. The overall or local characteristics of an image should be carefully emphasized. Original unclear images should be made clear or certain features of interest should be emphasized. The differences in the characteristics of different objects in an image should be enlarged, and the features that are not of interest should be suppressed. The image quality can be improved, its information can be enriched, image interpretation and recognition can be enhanced, and the practical application value of images can be increased [1]. The commonly used image enhancement methods can be divided into two main categories: Spatial and frequency domain enhancements. The former directly deals with the gray value of an image.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
