Detection of Blobs Using Variable Windows
Machi Saitoh* and Hiroshi Katsulai**
*Software Kogyo, Ltd., 2-9-2 Kanda Suda-cho, Chiyoda, Tokyo, 101 Japan
**Chiba Institute of Technology, 2-17-1 Tsudanuma, Narashino, Chiba, 275 Japan
A blob is a compact region in the image which is surrouded by a smoothly curved closed edge and, at the same time, is lighter or darker than the background. Many objects which are found in an image can be viewed as blobs. It is very important to quickly detect in an image the location of blobs for a wide range of applications of image analysis. In this paper, we propose a noniterative simple method of blob detection which is applicable to various forms of blobs by scanning an image with a size variable window, and does not require so much memory and time for computations because of its simplicity, and demonstrate its effectiveness through experiments.