|
Method development for verif ication the completeancient statues by image processing |
|---|---|
| รหัสดีโอไอ | |
| Creator | 1. Natthariya Laopracha 2. Umaporn Saisangjan 3. Rapeeporn Chamchong |
| Title | Method development for verif ication the completeancient statues by image processing |
| Publisher | Faculty of Engineering, Khon Kaen University |
| Publication Year | 2558 |
| Journal Title | KKU Engineering Journal |
| Journal Vol. | 42 |
| Journal No. | 2 |
| Page no. | 203-210 |
| Keyword | Image processing,Statue detection,Feature extraction,Object detection |
| ISSN | 0125-8273 |
| Abstract | Ancient statues are cultural heritages that should be preserved and maintained. Nevertheless, such invaluable statues may be targeted by vandalism or burglary. In order to guard these statues by using image processing, this research aims to develop a technique for detecting images of ancient statues with missing parts using digital image processing. This paper proposed the effective feature extraction method for detecting images of damaged statues or statues with missing parts based on the Histogram Oriented Gradient (HOG) technique, a popular method for object detection. Unlike the original HOG technique, the proposed method has improved the area scanning strategy that effectively extracts important features of statues. Results obtained from the proposed method were compared with those of the HOG method. The tested image dataset composed of 500 images of perfect statues and 500 images of statues with missing parts. The experimental results show that the proposed method yields 99.88% accuracy while the original HOG method gives the accuracy of only 84.86%. |
