| Budget Amount *help |
¥3,900,000 (Direct Cost: ¥3,000,000、Indirect Cost: ¥900,000)
Fiscal Year 2013: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000)
Fiscal Year 2012: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Fiscal Year 2011: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000)
|
|---|
| Research Abstract |
In this research, a new image measurement method has been proposed. By using a small number of image information, our method has achieved high speed image matching and object recognition. Its processing time is approximately 1000 times shorter than conventional methods.
I have shown that useful information for image matching is unique pixels and it can be calculated by estimation of probability of co-occurrence of pixel intensity. Robustness for varying illumination was performed by measuring spatial and temporal co-occurrence behaviors. In addition, stable classification ability has been realized by estimating distinguishability of the data for similar objects around the target object. Moreover, I have proposed additional methods to apply this basic idea to 3D object recognition and robust matching for partial occlusion.
I have proved that the proposed method can realize high-speed and reliable image matching by setting appropriate strategy for reducing amount of data.
|
