Limits...
A Multispectral Image Creating Method for a New Airborne Four-Camera System with Different Bandpass Filters.

Li H, Zhang A, Hu S - Sensors (Basel) (2015)

Bottom Line: For this multispectral system, an automatic multispectral data composing method was proposed.For the difficult registration problem between visible band images and near-infrared band images in cases lacking manmade objects, we presented an effective method based on the structural characteristics of the system.Experiments show that our method can acquire high quality multispectral images and the band-to-band alignment error of the composed multiple spectral images is less than 2.5 pixels.

View Article: PubMed Central - PubMed

Affiliation: Key Laboratory of 3D Information Acquisition and Application of Ministry, Capital Normal University, Beijing 100048, China. lihanlun@126.com.

ABSTRACT
This paper describes an airborne high resolution four-camera multispectral system which mainly consists of four identical monochrome cameras equipped with four interchangeable bandpass filters. For this multispectral system, an automatic multispectral data composing method was proposed. The homography registration model was chosen, and the scale-invariant feature transform (SIFT) and random sample consensus (RANSAC) were used to generate matching points. For the difficult registration problem between visible band images and near-infrared band images in cases lacking manmade objects, we presented an effective method based on the structural characteristics of the system. Experiments show that our method can acquire high quality multispectral images and the band-to-band alignment error of the composed multiple spectral images is less than 2.5 pixels.

No MeSH data available.


The dotted line shows a false match because pi is too far away from (x + xrow, y + ycol). And, the match showed by the solid line will be remained.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4541943&req=5

sensors-15-17453-f003: The dotted line shows a false match because pi is too far away from (x + xrow, y + ycol). And, the match showed by the solid line will be remained.

Mentions: If the flying height is fixed, there are no changes in flight attitude and no ups and downs on the ground, the camera plane parallels to the ground surface, the cameras are arrangement in parallel, there is no camera lenses distortion and CCD distortion, then the displacement (xrow, ycol) between the two images is fixed. If all of the dx, M, d, H and f are known, the (xrow, ycol) can be calculated by using Equation (10) directly. However, if not all of these parameters are known, we need other methods. An effective histogram method is proposed, and we will describe it in Section 4. For any feature point p0 with coordinate (x, y) in reference image I0, the coordinate of its matching point pi is (x + xrow, y + ycol) in input image Ii. Although these ideal situations do not occur in actuality, the influence of all these factors is limited for our multispectral system because the arrangement of the four cameras is almost parallel, as Figure 1. So, it can be estimated that pi is near (x + xrow, y + ycol) in the input image Ii. We just need to set a threshold to check whether pi is near (x + xrow, y + ycol). If the threshold is too low, some correct matches can be rejected. And, if it is too high, the removed false matches will decrease. In this paper, the threshold is set to one tenth of the image size, 104 pixels. If the Euclidean distance between pi and (x + xrow, y + ycol) is lower than this threshold, the match will be retained in this step, described as the solid line in Figure 3; otherwise, it will be taken as a false matching point and be eliminated, described as the dotted line in Figure 3. Due to this, the correct rate of initial matches will increase significantly and the RANSAC approach will become more reliable.


A Multispectral Image Creating Method for a New Airborne Four-Camera System with Different Bandpass Filters.

Li H, Zhang A, Hu S - Sensors (Basel) (2015)

The dotted line shows a false match because pi is too far away from (x + xrow, y + ycol). And, the match showed by the solid line will be remained.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4541943&req=5

sensors-15-17453-f003: The dotted line shows a false match because pi is too far away from (x + xrow, y + ycol). And, the match showed by the solid line will be remained.
Mentions: If the flying height is fixed, there are no changes in flight attitude and no ups and downs on the ground, the camera plane parallels to the ground surface, the cameras are arrangement in parallel, there is no camera lenses distortion and CCD distortion, then the displacement (xrow, ycol) between the two images is fixed. If all of the dx, M, d, H and f are known, the (xrow, ycol) can be calculated by using Equation (10) directly. However, if not all of these parameters are known, we need other methods. An effective histogram method is proposed, and we will describe it in Section 4. For any feature point p0 with coordinate (x, y) in reference image I0, the coordinate of its matching point pi is (x + xrow, y + ycol) in input image Ii. Although these ideal situations do not occur in actuality, the influence of all these factors is limited for our multispectral system because the arrangement of the four cameras is almost parallel, as Figure 1. So, it can be estimated that pi is near (x + xrow, y + ycol) in the input image Ii. We just need to set a threshold to check whether pi is near (x + xrow, y + ycol). If the threshold is too low, some correct matches can be rejected. And, if it is too high, the removed false matches will decrease. In this paper, the threshold is set to one tenth of the image size, 104 pixels. If the Euclidean distance between pi and (x + xrow, y + ycol) is lower than this threshold, the match will be retained in this step, described as the solid line in Figure 3; otherwise, it will be taken as a false matching point and be eliminated, described as the dotted line in Figure 3. Due to this, the correct rate of initial matches will increase significantly and the RANSAC approach will become more reliable.

Bottom Line: For this multispectral system, an automatic multispectral data composing method was proposed.For the difficult registration problem between visible band images and near-infrared band images in cases lacking manmade objects, we presented an effective method based on the structural characteristics of the system.Experiments show that our method can acquire high quality multispectral images and the band-to-band alignment error of the composed multiple spectral images is less than 2.5 pixels.

View Article: PubMed Central - PubMed

Affiliation: Key Laboratory of 3D Information Acquisition and Application of Ministry, Capital Normal University, Beijing 100048, China. lihanlun@126.com.

ABSTRACT
This paper describes an airborne high resolution four-camera multispectral system which mainly consists of four identical monochrome cameras equipped with four interchangeable bandpass filters. For this multispectral system, an automatic multispectral data composing method was proposed. The homography registration model was chosen, and the scale-invariant feature transform (SIFT) and random sample consensus (RANSAC) were used to generate matching points. For the difficult registration problem between visible band images and near-infrared band images in cases lacking manmade objects, we presented an effective method based on the structural characteristics of the system. Experiments show that our method can acquire high quality multispectral images and the band-to-band alignment error of the composed multiple spectral images is less than 2.5 pixels.

No MeSH data available.