Abstract
Abstract. Due to the raw images of multi-lens multispectral (MS) camera has significant misregistration errors, performing image registration for band co-registration is necessary. Image matching is an essential step for image registration, which obtains conjugate features on the overlapped areas, and use them to estimate the coefficients of a transformation model for correcting the geometrical errors. However, due to the none-linear intensity of spectral response, performing feature-based image matching (such as SURF) can only obtain only a few conjugate features on cross-band MS images. Different to SURF that extracts local extremum in a multi-scale space and utilizes a threshold to determine a feature, we proposed a normalized SURF (N-SURF) that extracts features on single scale, calculates the cumulative distribution function (CDF) of features, and obtains consistent features from the CDF. In this study, two datasets acquired from Tetracam MiniMCA-12 and Micasense RedEdge Altum are used for evaluating the matching performance of N-SURF. Results show that N-SURF can extract approximately 2–3 times number of features, match more points, and have more efficient than original SURF. On the other hand, with the successful of MS image matching, we can therefor use the conjugates to compute the coefficients of a geometric transformation model. In this study, three transformation models are used to compare the difference on MS band co-registration, i.e. affine, projective, and extended projective. Results show that extended projective model is better than the others as it can compensate the difference of lens distortion and viewpoint, and has co-registration accuracy of 0.3–0.6 pixels.
Highlights
Multi-lens multispectral (MS) imaging system adopts highly synchronized multiple lenses, which can use different sensor and spectral filter combinations to obtain different image resolution, different wavelength, and different number of MS images
In this study, based on SpeededUp Robust Features (SURF) features, we proposed a normalized SURF (N-SURF) that can extract more interest points and has ability to control the number of points on each image
First is to introduce the difference of N-SURF and original SURF, and utilize three evaluation indexes, i.e. Matching Rate (MS), Duplicate Rate (DR), and Correct Rate (CR), to evaluate their performances on MS image matching
Summary
Multi-lens multispectral (MS) imaging system adopts highly synchronized multiple lenses, which can use different sensor and spectral filter combinations to obtain different image resolution, different wavelength, and different number of MS images. Some state-of-the-art multi-lens MS cameras and their specifications can be found respectively in Figure 1 and Table 1, which contains one twelve lenses (Tetracam MiniMCA-12) and one five lenses (Micasense RedEdge Altum) MS camera. Both cameras can acquire blue (BLU), green (GRE), red (RED), rededge (REG), and near infrared (NIR) spectral response. Due to the slightly difference of perspective centers and viewing angles, the raw images of multi-lens MS camera have significant misregistration errors. Since multi-lens MS cameras are getting attraction on the applications of precision agriculture and environmental monitoring (Mulla, 2013; TorresSanchez et al, 2013), performing image matching and image registration (Mulla, 2013; Torres-Sanchez et al, 2013) to achieve the goal of band co-registration is important
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
