An Automatic Image Stitching Method for Visual Inspection of Concrete Structures
| Project/Area Number |
24760364
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (B)
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Structural engineering/Earthquake engineering/Maintenance management engineering
|
|---|
| Research Institution | Yamaguchi University |
|---|
Principal Investigator |
FUJITA Yusuke 山口大学, 医学(系)研究科(研究院), 准教授 (40509527)
|
|---|
| Project Period (FY) |
2012-04-01 – 2014-03-31
|
| Project Status |
Completed (Fiscal Year 2013)
|
| Budget Amount *help |
¥3,510,000 (Direct Cost: ¥2,700,000、Indirect Cost: ¥810,000)
Fiscal Year 2013: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2012: ¥2,470,000 (Direct Cost: ¥1,900,000、Indirect Cost: ¥570,000)
|
|---|
| Keywords | 画像処理 / コンクリート構造物 / 維持管理 / 画像合成 / ひび割れ / 外観検査 |
|---|
| Research Abstract |
In this study, an image stitching method for effective visual inspection of the concrete structure is developed. At first, a self-calibration method using straight lines on concrete surfaces was proposed. Even if the focal distance of an image is different from that of other images, lens distortion are corrected automatically before image stitching. Further, images of multiple views are used for image stitching, to acquire a composite image having a supplemented dead angle region part. Thus, the dead angle area of an overview image is canceled and the operator can recognize the damages such as cracks within the dead angle area of the overview image. After that, the contrasts of each pixel existing in the same position between the synthesized image data are compared with each other, to take out the pixel whose contrast is maximum and create an overview image data, to improve contrast the composite image.
|
Report
(3 results)
Research Products
(4 results)
