A digital auto-focusing method based on CCD mosaicing for aerial camera

Abstract

A novel focusing method of the remote sensing camera is proposed in this paper. To evaluate the quality of the image obtained by the aerial camera, an assessing function was constructed based on the Wavelet transform. And the characteristic of the CCD mosaicing structure was taken use to solve the problem that the evaluating values can not be compared for the images captured by the aerial camera are variable. On the basis of the wavelet evaluating function, the quality of the image was assessed. Then the CCD mosaicing structure was utilized to perform auto-focusing process, a simulation was made to validate the novel method in the end. There are three major contributions in the paper. Firstly, the weights of the wavelet coefficients in the evaluating function were set according to the characteristic of the Wavelet transform and human vision system (HVS). By those, the assessing result is close to our subjective feel and insensitive to noise. Secondly, in order to make the function adaptive to images with different high-frequency components, the properties of the wavelet basis were analyzed. By comparing the evaluating effect of different images, the wavelet basis with the best effect is symlet2 and the level of decomposing is three. Finally, by making use of the CCD mosaicing structure, the problem that auto-focusing of the aerial camera can't use digital images processing directly was solved, and the region with the highest frequency component was chosen as the evaluating area so as to improve the sensitivity of the function.

Analysis of image motion on autofocus precision for aerial cameras

“…In addition to some well-known sharpness measure functions, 9,15 frequency-based focus measures have been proposed in recent years. 8,13,16 They transform the image from spatial domain into frequency domain via a Fourier transform or a wavelet transform.…”

“…[1][2][3][4][5][6][7][8][9][10] The image-based passive autofocus algorithms are widely studied for their high accuracy and no requirement for extra equipment. [1][2][3][4][5][6][7][8][9][10] The image-based passive autofocus algorithms are widely studied for their high accuracy and no requirement for extra equipment.…”

“…[3][4][5][6][7][8][9][10] To date, there has been no theoretical investigation of the relationship between image motion and autofocus techniques. [3][4][5][6][7][8][9][10] To date, there has been no theoretical investigation of the relationship between image motion and autofocus techniques.…”