Improving Disparity Map of A Specific Object in A Stereo Image Using Camera Calibration, Image Rectification, and Object Segmentation

Syahputra, Hermawan (2014) Improving Disparity Map of A Specific Object in A Stereo Image Using Camera Calibration, Image Rectification, and Object Segmentation. International Journal of Applied Engineering Research (IJAER), 9 (22). pp. 15309-15319. ISSN 0973-4562

	Text Fulltext.pdf - Published Version Download (2MB)
	Text Similaritas.pdf - Published Version Download (3MB)
	Text Reviewer.pdf - Published Version Download (882kB)

Abstract

The objective of the study was to improve the disparity map of a certain object in a stereo image. The steps required to complete this are first, to perform camera calibration in order to extract camera parameters, and then to use these parameters to rectify the stereo image. Second, to perform segmentation to separate the object from the background of the image. Separation of the object from the background was performed to produce a good disparity for the object. This had not been done in previous studies. Using segmentation, bias and error of stereo correspondence had been reduced, since the correspondence was focus on the intensity of the left and right objects in the stereo image. Using rectified and segmented stereo images, better disparity maps had been obtained.

Actions (login required)

View Item