标题：Performance Comparison of Density Distribution and Sector mean of sal and cal functions in Walsh Transform Sectors as Feature Vectors for Image Retrieval
期刊名称：International Journal of Image Processing (IJIP)
电子版ISSN：1985-2304
出版年度：2010
卷号：4
期号：3
页码：205-217
出版社：Computer Science Journals
摘要：In this paper we have proposed two different approaches for feature vector generation with absolute difference as similarity measuring parameter. Sal-cal vectors density distribution and Individual sector mean of complex Walsh transform. The cross over point performance of overall average of precision and recall for both approaches on all applicable sectors sizes are compared. The complex Walsh transform is conceived by multiplying sal components by j= ã-1. The density distribution of real (cal) and imaginary (sal) values and individual mean of Walsh sectors in all three color planes are considered to design the feature vector. The algorithm proposed here is worked over database of 270 images spread over 11 different classes. Overall Average precision and recall is calculated for the performance evaluation and comparison of 4, 8, 12 & 16 Walsh sectors. The overall average of cross over points of precision and recall is of all methods for both approaches are compared. The use of Absolute difference as similarity measure always gives lesser computational complexity and Individual sector mean approach of feature vector has the best retrieval.
关键词：CBIR; Precision and Recall; Eucledian Distance; Kekre's Algorithm; Walsh Transform
