| 摘要 |
PROBLEM TO BE SOLVED: To prevent image data in an ineffective area from being included in the extraction of feature value due to a setting mistake by limiting the processing object area of the image data to an effective are via a feature value extraction means. SOLUTION: The image data on an image input means 101 such as a CCD camera are inputted to a 1st image 100 and the effective areas of the image 100 are stored in a 1st effective area storage means 102. Meanwhile, the image 100 is processed by an image processing means 103 and a 2nd image 104 is generated. Also, a feature value extraction means 107 extracts a pixel mean value, a standard deviation, a minimum value, a maximum value, a gradation histogram, a projection distribution, etc., from the image 104. Then, a processing object area is automatically limited to an effective area that is decided by an effective area decision means 105. Thus, it's possible to evade such a case that the image data on an in effective area are included in extraction of feature value due to a setting mistake.
|
