自适应直方图均衡化
散斑噪声
人工智能
反锐化掩蔽
斑点图案
计算机科学
计算机视觉
超声波
对比度(视觉)
各项异性扩散
直方图均衡化
图像处理
降噪
图像(数学)
医学
放射科
作者
Rika Favoria Gusa,Risanuri Hidayat,Hanung Adi Nugroho
标识
DOI:10.1109/iaict59002.2023.10205792
摘要
Ultrasound imaging is widely used in medical diagnosis because it is non-invasive and free from ionizing radiation. However, ultrasound images have low contrast and contain speckle noise, making diagnosis difficult. Therefore, speckle noise reduction and image contrast enhancement are important prerequisites in ultrasound image processing. Many methods can be used in the ultrasound image pre-processing stage. In this paper, fetal ultrasound images were enhanced in contrast and sharpness using four enhancement methods, namely histogram equalization (HE), contrast limited adaptive histogram equalization (CLAHE), unsharp masking (UM), and maximum local variation-based unsharp masking (MLVUM). These methods were applied to ultrasound images in two ways. Those are without filtering them and by first filtering them using a speckle reducing anisotropic diffusion (SRAD) filter. A comparative analysis was carried out on the performance of the four enhancement methods using the absolute mean brightness error (AMBE), average local contrast (ALC), and average gradient (AG) parameters. The results show that UM and MLVUM work better in increasing the contrast of fetal ultrasound images than HE and CLAHE. Applying the HE, CLAHE, UM, and MLVUM methods without filtering produces ultrasound images with better sharpness and contrast than enhanced images involving filtering but causing speckle noise amplification.
科研通智能强力驱动
Strongly Powered by AbleSci AI