Translated Title
Extraction algorithm of least square method based on fish-eye lens contour
Translated Abstract
Although with the advantages of super wide angle in the fisheye image ,we can’t ignore large visual distortion from the image .Fisheye distorted image contour extraction is the premise of correcting picture .With the method of least squares fitting circular contour ,we introduce the use of iterative correction method to calculate the center and radius of the data points on the edge of the filter ,we use feedback link to amend center and radius in order to ensure the accuracy of data .Also in this paper ,when there are large reflective areas in the invalid lens area ,we proposed a neighborhood weights method to generate the binary image which could suppress reflex areas outside the region .Experiments show that this method has high precision ,good stability ,and strong adaptability .
Translated Keyword
the least square method
fisheye lens
contour extraction
neighborhood weights
Access Number
WF:perioarticalgwdzcljs201504008
Corresponding authors email