当前位置: X-MOL 学术J. Ambient Intell. Human. Comput. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Biomedical image compression using fuzzy transform and deterministic binary compressive sensing matrix
Journal of Ambient Intelligence and Humanized Computing Pub Date : 2020-05-19 , DOI: 10.1007/s12652-020-02103-x
S. Nirmalraj , G. Nagarajan

Bio medical images are very important in analysing the internal structure of the body in a non-invasive method, to diagnose for any abnormalities and provide proper treatment to the patients. Medical images include X-rays, CT scan and MRI scan etc. It is necessary to compress these images, so that they can be easily stored and communicated from one place to another. The result of compression should be in such a way that, quality of the compressed image should be good and the compression ratio should be high. The biomedical images are mainly prone to impulse noise which may lead to degradation of image quality. So it is important to filter the impulse noise from the biomedical images without affecting the details of the image. In this paper the filtering of impulse noise from the biomedical images is done using fuzzy transform, followed by compressive sensing to compress the bio medical images without losing information content. Compressive sensing uses a sensing matrix to measure random samples from the image signal. This paper proposes the compression of biomedical images using a deterministic binary compressive sensing matrix and for the recovery of the biomedical image, Orthogonal Matching Pursuit (OMP) is used.



中文翻译:

基于模糊变换和确定性二进制压缩感知矩阵的生物医学图像压缩

生物医学图像对于以非侵入性方法分析身体内部结构,诊断任何异常情况并为患者提供适当治疗非常重要。医学图像包括X射线,CT扫描和MRI扫描等。必须压缩这些图像,以便可以轻松地将它们存储并从一个地方传递到另一个地方。压缩的结果应以这样的方式进行:压缩图像的质量应良好,压缩率应较高。生物医学图像主要易于产生脉冲噪声,这可能导致图像质量下降。因此,重要的是从生物医学图像中滤除脉冲噪声,而又不影响图像的细节。本文使用模糊变换对生物医学图像中的脉冲噪声进行过滤,然后进行压缩感测以压缩生物医学图像而不会丢失信息内容。压缩感测使用感测矩阵来测量图像信号中的随机样本。本文提出使用确定性二进制压缩感知矩阵对生物医学图像进行压缩,并使用正交匹配追踪(OMP)来恢复生物医学图像。

更新日期:2020-05-19
down
wechat
bug