DocumentCode :
238504
Title :
Comparative analysis of demosaicking techniques for Mars Colour Camera data
Author :
Roy, Sandip ; Dhar, Debajyoti ; Moorthi, S.M. ; Sarkar, S.S.
Author_Institution :
Space Applic. Centre, MSDPD/DPSG/SIPA, ISRO, Jodhpur, India
fYear :
2014
fDate :
27-29 Nov. 2014
Firstpage :
862
Lastpage :
867
Abstract :
A digital camera captures the light that falls onto the sensor. Imaging begins with the source of light called illumination and all illuminations virtually consist of light with multiple wavelengths. The captured object can be represented either as panchromatic image or multispectral image. In multispectral imaging, each pixel holds an intensity values different for different spectral resolution. In this imaging multiple bands are realized using field splitting technique near focal plane. The final selection of the spectral bands is achieved by using appropriate band pass filters in front of the detectors. But with the advancement of time and technology in order to reduce cost and size, single sensor with color filter arrays (CFA) has introduced and improved. The most common array is the Bayer color filter array, where the green image is measured at a higher sampling rate because the peak sensitivity of the human visual system lies in the medium wavelengths, corresponding to the green portion of the spectrum. Most of the digital sensors are “Bayer Sensor”. A few digital cameras have a relatively new type of sensor that use the principle of depth of penetration and capture three color values at each pixel. In case of Bayer, individual pixel will have single spectral information at a time depending on the colour filter of the pixel. The reconstruction of the full color image requires each pixel must have R, G and B values, however typically only one color is sampled at a particular spatial location. A subsequent interpolation step is needed which is commonly called “Demosaicking” [2]. The process of demosaicking introduces various artifacts commonly false coloring, zippering effect. These artifacts are very less in inter-channel interpolation, but it changes the output values of the RGB channels. In Data Processing chain visual quality and value of the corresponding digital number (DN) which is being used in physical unit (radiance) co- version should not be deviated much from the original raw data. These three methods are taken from the broader categories of demosaicking. Comparative analysis is carried out.
Keywords :
cameras; colour photography; digital photography; focal planes; light sources; Bayer sensor; Demosaicking technique; Mars colour camera data; band pass filters; digital camera; digital sensors; field splitting technique; focal plane; full color image; light source; multispectral image; panchromatic image; spectral bands; Cameras; Channel estimation; Image color analysis; Image reconstruction; Image resolution; Interpolation; AP; Bilinear; Demosaicking; EBP; LMMSE; MCC; MOM;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Contemporary Computing and Informatics (IC3I), 2014 International Conference on
Conference_Location :
Mysore
Type :
conf
DOI :
10.1109/IC3I.2014.7019598
Filename :
7019598
Link To Document :
بازگشت