华南理工大学学报(自然科学版)
華南理工大學學報(自然科學版)
화남리공대학학보(자연과학판)
JOURNAL OF SOUTH CHINA UNIVERSITY OF TECHNOLOLGY
2002年
1期
17-21
,共5页
图像处理%模式识别%人脸识别%灰度分布%标准化
圖像處理%模式識彆%人臉識彆%灰度分佈%標準化
도상처리%모식식별%인검식별%회도분포%표준화
image processing%pattern recognition%face recognition%greyscale distribution%normalization
分析了传统的图像灰度分布的标准化算法,针对其缺陷进行了一些改进,并归并为"聚集算法".提出了这些算法的评价方法,并通过实验结果证实了算法改进方式的合理性.
分析瞭傳統的圖像灰度分佈的標準化算法,針對其缺陷進行瞭一些改進,併歸併為"聚集算法".提齣瞭這些算法的評價方法,併通過實驗結果證實瞭算法改進方式的閤理性.
분석료전통적도상회도분포적표준화산법,침대기결함진행료일사개진,병귀병위"취집산법".제출료저사산법적평개방법,병통과실험결과증실료산법개진방식적합이성.
In this paper, the traditional normalization algorithm of image greyscale distribution is analyzed and some improvements according to its shortcomings are made. These improvements are added to the traditional theory, then a new theory called “congregation algorithm” is established. Methods of evaluations for these algorithms are also designed and the experimental results support the rationality of the improvements.