CAJ | 학술논문

论文通过多尺度分析小波变换，探索一种肝脏 B 超图像噪声滤除方法，提高脂肪肝 B 超图像识别的准确率。论文提出一种基于小波变换多尺度分析的改进阈值噪声滤除方法。首先建立散斑噪声模型，然后进行多尺度小波变换，利用改进的自适应阈值滤噪方法去除图像噪声，最后经小波逆变换重构肝脏 B 超图像。实验结果显示肝脏 B 超原始图像中的散斑噪声被有效滤除，同时较好地保存了肝脏 B 超图像的细节信息。该方法可有效地滤除肝脏 B 超图像中的散斑噪声，较好地保留细节信息，提高了脂肪肝 B 超图像识别的准确率。
논문통과다척도분석소파변환，탐색일충간장 B 초도상조성려제방법，제고지방간 B 초도상식별적준학솔。논문제출일충기우소파변환다척도분석적개진역치조성려제방법。수선건립산반조성모형，연후진행다척도소파변환，이용개진적자괄응역치려조방법거제도상조성，최후경소파역변환중구간장 B 초도상。실험결과현시간장 B 초원시도상중적산반조성피유효려제，동시교호지보존료간장 B 초도상적세절신식。해방법가유효지려제간장 B 초도상중적산반조성，교호지보류세절신식，제고료지방간 B 초도상식별적준학솔。
A method of B-Scan ultrasonic images denoising based on wavelet transform is explored .It improved accura-cy of B-Scan ultrasonic image recognition of fatty liver .A speckle suppression method for B-Scan ultrasound image based on improved threshold of wavelet transform is presented .Firstly ,a speckle noise model that is carried out to the B-Scan ultra-sound image is built .Secondly ,the image is decomposed into low frequency and high frequency information by wavelet trans-form .Adaptive threshold denoising are carried out to the fatty liver B-Scan ultrasonic images .Finally ,the inverse wavelet transform is processed .The speckle noises in the original image are removed efficiently and the image edge details are re-served .The method is denoising speckle noises and reserving edge details in B-Scan ultrasonic image effciently .It improves accuracy of B-Scan ultrasonic image recognition of fatty liver .